1

Ecological and economic benefits of the application of 1

bio-based mineral fertilizers in modern agriculture 2 3 4C. Vaneeckhautea, E. Meersa, E. Michelsa, J. Buysseb, F.M.G. Tacka 5

6 7

8a C. Vaneeckhaute, E. Meers, E. Michels, F.M.G. Tack 9 Laboratory of Analytical and Applied Ecochemistry, 10 Faculty of Bioscience Engineering, 11 University of Ghent 12 Coupure Links 653, 9000 Ghent, Belgium 13 E-mail: Celine.Vaneeckhaute@ugent.be, Erik.Meers@ugent.be, Evi.Michels@ugent.be, 14 Filip.Tack@ugent.be 15 16 17b J. Buysse 18 Department of Agricultural Economics, 19 Faculty of Bioscience Engineering, 20 University of Ghent 21 Coupure Links 653, 9000 Ghent, Belgium 22 E-mail: Jeroen.Buysse@ugent.be 23 24 25Corresponding author: 26

a C. Vaneeckhaute 27 Laboratory of Analytical and Applied Ecochemistry, 28 Faculty of Bioscience Engineering, 29 University of Ghent 30 Coupure Links 653, 9000 Ghent, Belgium 31 E-mail: Celine.Vaneeckhaute@ugent.be 32 Tel.: +32/478.43.24.64 33 Fax: +32/9.264.62.42 34 35 36 37

38

39

40

41

2

Abstract 42

In the transition from a fossil to a bio-based economy, it has become an important challenge 43

to maximally recuperate valuable nutrients coming from waste streams. Nutrient resources are 44

rapidly depleting, significant amounts of fossil energy are used for the production of chemical 45

fertilizers, whereas costs for energy and fertilizers are increasing. In the meantime, biogas 46

production through anaerobic digestion produces nutrient-rich digestates. In high-nutrient 47

regions, these products cannot or only sparingly be returned to agricultural land in its crude 48

unprocessed form. The consequent processing of this digestate requires a variety of 49

technologies producing lots of different derivatives, which could potentially be re-used as 50

green fertilizers in agriculture. As such, a sustainable alternative for fossil-based mineral 51

fertilizers could be provided. This study aims to characterize the physico-chemical properties 52

of digestates and derivatives, in order to identify the fertilizer value and potential bottlenecks 53

for agricultural re-use of these products, in line with European legislative constraints. In 54

addition, the economic and ecological benefits of substituting conventional fertilizers by bio-55

based alternatives are quantified and evaluated. Waste water from acidic air scrubbers for 56

ammonia removal shows potential for application as N-S fertilizer. Analogously, concentrates 57

resulting from membrane filtrated liquid fraction of digestate show promise as N-K fertilizer. 58

Substituting conventional fertilizers by digestate derivatives in different cultivation scenarios 59

can result in significant economic and ecological benefits for the agriculturist. Starting from 60

theoretical scenarios outlined in the current study, field test validation will be required to 61

confirm the potential substitution of fossil-based mineral fertilizers by bio-based alternatives. 62

63

64

65Abbreviations: AF = Artificial Fertilizer, AM = Animal Manure, DD = Digestate Derivatives, LF = Liquid Fraction, NUE

= Nitrogen Uptake Efficiency, TF = Thick Fraction

3

Introduction 66

In 2010, chemical fertilizer use in Europe (EU-27) was as high as 10.4 Mt of nitrogen (N), 2.4 67

Mt of phosphate (P2O5) and 2.7 Mt of potash (K2O) [1]. By 2019/2020, forecasters expect 68

these fertilizer consumption figures to reach 10.8 Mt, 2.7 Mt and 3.2 Mt, respectively [1]. 69

Unfortunately, fertilizer production requires significant amounts of fossil energy. Up to 29 GJ 70

t-1 is used for the production of NH4 through the Haber-Bosch process under optimal 71

conditions [2]. Also, prices for mineral fertilizers are increasing, whereas nutrient resources 72

are depleting [3], [4], [5], and [6]. Particularly phosphorus is a nutrient with fast increasing 73

scarcity. It is expected that the direct available phosphorus resources will be completely 74

depleted by the end of this century [5], [6], and [7]. In the transition from a fossil-based to a 75

bio-based economy, it has therefore become an important challenge to maximally recuperate 76

and recycle valuable nutrients from waste streams in a sustainable and environmentally 77

friendly manner. 78

In frame of the 2020 directives, the conversion of biomass such as energy crops, organic 79

residues and animal wastes, into biogas through anaerobic digestion has been evaluated as one 80

of the most energy-efficient and environmentally beneficial technologies for bio-energy 81

production [8]. However, the anaerobic digestion of biomass produces besides renewable 82

energy also nutrient-rich digestates as a “waste” stream. In high-nutrient regions such as 83

Flanders (Belgium), Barcelona (Spain), Nord-Rein Westfalen (Germany), Bretagne (France), 84

Denmark and the Netherlands, these digestates need to be processed further and cannot or 85

only sparingly be returned to arable land as a fertilizer in its crude unprocessed form [9]. The 86

underlying reason for this technical prerequisite is that, due to intensive industrial animal 87

production, these regions are characterized by an overproduction of animal manure in 88

comparison to the available arable land. 89

Initial steps of digestate processing generally comprise the use of separation and/or 90

4

dewatering technologies, using emulsion or powder based polymers for flocculation [10]. The 91

resulting thick fractions are mostly dried, turning them into pasteurized and stabilized 92

exportable organic soil conditioners, high in phosphorus and organic matter [11]. The liquid 93

fraction (LF) produced by the separation step contains the majority of the digestate’s 94

potassium and inorganic nitrogen. This liquid fraction can be processed further by ammonia 95

stripping or membrane filtration, for example microfiltration (MF), ultrafiltration (UF) and/or 96

reversed osmosis (RO) [12] and [13]. Each step in a membrane cascade again generates two 97

downstream products, concentrate and permeate, with varying characteristics concerning 98

macro- and micronutrient composition. Alternatively, the liquid fraction after separation can 99

be treated biologically, for example by nitrification-denitrification. However, nitrification-100

denitrification ultimately converts valuable nitrogen into nitrogen gas (N2), which is then 101

eliminated from the local agricultural cycle. In regions where agricultural nitrogen emissions 102

to the environment are already excessive and in conflict with the European nitrate directive 103

for protection of water bodies, this nitrogen elimination may be economically and 104

ecologically sensible for a given portion of the nitrogen. Furthermore, exhaust gases of the 105

biogas cogen-engines and driers need to be washed before emission into the atmosphere. This 106

involves the use of acidic, alkaline and/or oxidative scrubber techniques, again resulting in 107

different types of specific streams, some of which contain large amounts of inorganic 108

nutrients. In acidic air scrubbers, for example, sulfuric acid is dosed to capture NH3 and 109

amines, thereby producing (NH4)2SO4 as a nutrient-rich waste stream. Alkaline air scrubbers 110

are then used to neutralize acidic components that escaped from the system or to oxidize 111

organic compounds such as H2S [14]. 112

The above described derivatives can function as either inorganic or organic fertilizers 113

and/or soil conditioners, in the meantime providing renewable substitutes for mineral 114

fertilizers based on fossil resources. Such a sustainable development strategy is in line with 115

5

the cradle-to-cradle approach [15]: waste turns into secondary resources (Fig. 1). 116

[Fig. 1 here] 117

In the past, only digestates in their crude form have been compared with animal manures 118

in comprehensive research [16]. Insights in the composition and properties of the more 119

important derivatives is lacking, though very relevant, as the treatment and the transport of 120

these products is expensive and energy consuming, while valuable nutrients are often wasted. 121

If a sustainable market for digestate and its derivatives would exist, the digestate problem 122

could be turned into an economic and ecological opportunity. Valorization of macro- and 123

micronutrients from digestate processing is therefore an important challenge for the future. 124

This study aims to characterize the physico-chemical properties of the different derivatives 125

coming from digestate processing, with attention for general conditions such as pH and 126

conductivity, macronutrients, essential and non-essential trace elements, organic carbon (OC) 127

and nutritive ratios. The fertilizer value and the potential bottlenecks for re-use of these 128

products in comparison with conventional fertilizers are identified. Finally, the economic and 129

ecological benefits of substituting conventional fertilizers by digestate derivatives are 130

calculated for the most relevant cultivation scenario’s. The knowledge obtained in this 131

research should greatly enhance the understanding and useful application of digestate and its 132

derivatives. Getting a better view on the dilemmas and opportunities posed by these products 133

can in turn help to improve the underlying economics of biodigestion. As such, this research 134

can serve as a catalyst to stimulate this vital, yet fragile, innovative economic activity in frame 135

of the 2020 objectives. 136

137

2. Material and methods 138

2.1 Site description and experimental set-up 139

Samples of the various digestate derivatives were taken in three different anaerobic digestion 140

6

plants in Belgium: Goemaere Eneco Energy Diksmuide, Mandel Eneco Energy Roeselare and 141

Sap Eneco Energy Merkem. The incoming feed to the digesters is composed of animal 142

manure, organic-biological waste and energy maize. The following process streams were 143

sampled: digestates, thick fractions (TF) of digestates after separation, thick fractions (TF) of 144

digestates after separation and drying, liquid fractions (LF) of digestates after separation, 145

concentrates produced by one vibrating membrane filtration step of the liquid fraction using 146

reversed osmosis (RO) membranes, concentrates produced by two subsequent vibrating 147

membrane filtration steps, waste water from an acidic air scrubber and waste water from an 148

alkaline air scrubber. For detailed process description, see [13]. The samples in the digestion 149

plants were taken on three different points in time during approximately one year (2010-150

2011). Acidic air scrubber water was additionally sampled two times at the pig farm of 151

Ladevo BVBA, Ruiselede, Belgium and conventional pig slurry was sampled two times at the 152

site of Huisman, Aalter, Belgium. The samples (10 L each) were collected in polyethylene 153

sampling buckets and transported within 1 h from the test site to the laboratory, carried in 154

cooler boxes filled with ice. In the laboratory, the replicate samples were stored cool (1–5 °C) 155

and kept separate for replicate analysis. 156

157

2.2 Liquid sample analysis 158

Conductivity and pH were determined potentiometrically using a WTW F537 conductivity 159

electrode (Wissenschaftlich Technischen Werkstäten, Weilcheim, Germany) and an Orion 160

model 520A pH meter (Orion Research, Boston, VS), respectively. Suspended solids (SS) 161

were determined by vacuum filtration (0.45 µm) of 100-300 mL sample and subsequent 162

drying of the filter in a furnace (Memmert, Schwabach, Germany) at 105 °C. Total nitrogen 163

content was determined using a Kjeltec system 1002 distilling unit (Gerhardt Vapodest, 164

Köningswinter, Germany) after digestion of the sample in a sulphuric-salicylic acid mixture. 165

7

Finally, the captured ammonia in the distillate was titrated with 0.01 mol L-1 HCl in the 166

presence of a methyl red bromocresol green mixed indicator [17]. Total phosphorus content 167

was determined using the colorimetric method of Scheel [17] after wet digestion of the liquid 168

samples (2.5 g sample + 2 mL HNO3 + 1 mL H2O2). The absorbance at 700 nm of samples 169

and standards was determined using a Jenway 6400 spectrophotometer (Barloworld Scientific 170

T/As Jenway, Felsted, UK). Calcium, magnesium and heavy metals were analyzed using ICP-171

OES (Varian Vista MPX, Palo Alto, CA, USA) after wet digestion (see above). Sodium and 172

potassium of the digested samples (see above) were analyzed using a flame photometer 173

(Eppendorf ELEX6361, Hamburg, Germany). Ammonium was determined using a Kjeltec 174

system 1002 distilling unit (Gerhardt Vapodest, Köningswinter, Germany) after addition of 175

MgO to the liquid sample (50 mL). Finally, the captured ammonia in the distillate was titrated 176

with 0.01 mol L-1 HCl in the presence of a methyl red bromocresol green mixed indicator 177

[17]. Nitrate, chloride and sulphate were analyzed using ionic chromatography (Metrohm 761, 178

Herisau, Switzerland) after centrifugation and subsequent vacuum filtration (0.45 µm) of the 179

liquid fraction. Total sulfur was analyzed as described by [18]. The extractable amount of 180

macronutrients was determined in an NH4OAc-EDTA pH 4.65 extract of the samples [17]. 181

182

2.3 Thick sample analysis 183

Dry weight (DW) content was determined as residual weight after 48 h drying at 100 °C. Ash 184

and organic carbon (OC) were determined by incineration of the dry samples in a furnace 185

(Nabertherm, Lilientahl, Germany) at 550 °C during 4 h. Conductivity and pH were measured 186

using a WTW F537 conductivity electrode (Wissenschaftlich Technischen Werkstäten, 187

Weilcheim, Germany) and an Orion model 520A pH meter (Orion Research, Boston, VS), 188

respectively, after equilibration for 1 h in deionized water at a 5/1 liquid to dry sample ratio 189

and subsequent filtering (white ribbon, MN 640 m, Macherey–Nagel, Düren, Germany). Total 190

8

nitrogen was determined using the Kjeldahl procedure on fresh weight (FW) content [17]. For 191

the determination of phosphorus, dry samples were incinerated at 450 °C during 4 h in a 192

furnace (Nabertherm, Lilientahl, Germany). The phosphorus content was then determined by 193

the colorimetric method of Scheel [17] after digestion of the residual ash (1 g ash + 5 mL 3 194

mol L-1 HNO3 + 5 ml 6 mol L-1 HNO3). Calcium and magnesium of the digested samples (see 195

above) were analyzed by means of ICP-OES (Varian Vista MPX, Palo Alto, CA, USA). 196

Sodium and potassium of the digested samples (see above) were determined using a flame 197

photometer (Eppendorf ELEX6361, Hamburg, Germany). Ammonium was determined using 198

a Kjeltec system 1002 distilling unit (Gerhardt Vapodest, Köningswinter, Germany) after 199

addition of MgO to the sample (50 mL). Finally, the captured ammonia in the distillate was 200

titrated with 0.01 mol L-1 HCl in the presence of a methyl red bromocresol green mixed 201

indicator [17]. Chloride was determined by means of a potentiometric titration using an 202

automatic titrator (Methrohm, Herisau, Switzerland), provided by a Hg/(Hg)2SO4 referential 203

electrode [17]. The available amount of macronutrients was determined in an NH4OAc-EDTA 204

pH 4.65 extract of the samples [17]. 205

206

2.4 Economic and ecological analysis 207

The economic and ecological benefits were calculated for the current most relevant re-use 208

scenarios of green fertilizers for the cultivation of maize (Table 1). 209

[Table 1 here] 210

For each scenario the total amount of available N applied to soil is 150 kg ha-1, according to 211

the Flemish manure regulation for the cultivation of maize on non-sandy soils [19]. Hereby, 212

the amount of available N in both animal manure and digestate derivatives was considered to 213

be 60 % of the total N content, as described by policy [19]. Furthermore, the maximum 214

application standard of 80 kg ha-1 y-1 for total P2O5 [19] and a total K2O dose of 220 kg ha-1 y-1 215

9

were respected for each scenario. The different scenarios are compared with the common 216

practice (scenario 0): maximum amount of animal manure (P2O5 = limiting factor), artificial 217

start fertilizer (N) and additional optimizing fertilization with artificial manure (N, K2O), with 218

respect for the maximum allowable levels of N and P2O5 application on agricultural land [19]. 219

The data used for the economic and ecological analysis of the different cultivation 220

scenarios are presented in Table 2. 221

[Table 2 here] 222

The economic and ecological impact of artificial fertilizer production, packing, transport and 223

application was taken in account. Hereby the energy use for transport and application was 224

calculated for a lorry with a capacity of 20 t and a diesel consumption of 11.6 MJ km-1 225

(personal communication, EnergieTransitie, 2011; [22]) that travels from the port of Antwerp, 226

distribution point in Belgium, to Ypres in the west of Flanders (129 km), region with the 227

highest nutrient use in Belgium. Next, the impact of transport and application of animal 228

manure and digestate derivatives were incorporated in the calculation [22]. Here it was 229

assumed that the transport distance from the farm to the field is less than 5 km and that a 230

tractor of 88.3 kW is used, with a diesel consumption of 10 L h-1. In this way, it is possible to 231

apply 30 t ha-1 h-1 of animal manure or digestate derivatives, which are common figures [22]. 232

The transport costs were then calculated based on the current average cost prize for diesel in 233

Europe (1.37 € L-1 [23]). Further, it was assumed that an agricultural contractor was paid 2.5 234

€ t-1 FW for fertilizer application [9]. 235

Next to these costs, also the economic benefits for the agriculturist when accepting animal 236

manure or digestate derivatives as base-fertilizer were handled. This amounts to 250 € ha-1 237

[9], resulting in 11.9 € t-1 FW or 1.47 € kg-1 N for animal manure. When animal manure is 238

substituted by digestate, LF digestate and a mixture of digestate (φ = 0.5) and LF digestate 239

(φ = 0.5), the benefits are 9.99 € t-1 FW, 5.29 € t-1 FW and 6.91 € t-1 FW respectively, based 240

10

on the N content of these streams. Nevertheless, it is expected that in the future these benefits 241

will have to be calculated using the P content of the product, in line with the legislative P 242

standards for soil application that become the more and more strict [19]. 243

Based on all these data, the economic and ecological impact was calculated using the 244

following functions: 245

Economic cost (€ ha-1) = AFproduction + AFpacking + AFapplication + AFtransport + DDapplication + 246

DDtransport + AMapplication + AMtransport – AM/DDbenefits (1) 247

Energy use (GJ ha-1) = AFproduction + AFpacking + AFtransport + AFapplication + DDtransport + 248

DDapplication + AMtransport + AMapplication (2) 249

where “AF” are artificial fertilizers, “DD” are digestate derivatives and “AM” is animal 250

manure. 251

Finally, when fossil-based mineral fertilizers are replaced by digestate derivatives, 252

significant savings in greenhouse gas (GHG) emissions can be expected. The GHG-emission 253

was calculated for the different scenarios in terms of carbon dioxide (CO2) equivalents (kg ha-254

1). It was assumed that diesel is used for the transport and application of fertilizers and that 255

natural gas is used for the production of artificial fertilizers. 256

257

3. Results 258

3.1 Physico-chemical analysis 259

Digestates, thick (TF) and liquid (LF) fractions of digestates after separation, thick fractions 260

(TF) of digestates after separation and drying, as well as conventional pig slurry were sampled 261

and physico-chemically analyzed (Table 3). Also, a mixture of digestate (φ = 0.5) and LF 262

digestate (φ = 0.5) was made and characterized (Table 3). 263

[Table 3 here] 264

11

The N/P/K ratio was very variable for the different products: 2.8/1/1.6 for digestates, 265

0.77/1/0.36 and 13/1/11 for thick and liquid fractions after separation respectively, 5.4/1/2.5 266

for the mixtures and 3.4/1/1.5 for pig slurry. Micronutrient contents were in all samples lower 267

than the Flemish legislation criteria [24] for use as fertilizer and/or soil conditioner in 268

agriculture. 269

Furthermore, concentrates produced by one vibrating membrane (RO) filtration step of LF 270

digestate, as well as concentrates following two subsequent vibrating membrane (RO) 271

filtration steps were sampled and physico-chemically analyzed (Table 4). 272

[Table 4 here] 273

Results show that concentrates produced by the first filtration not only contained more 274

macronutrients and organic carbon on Fresh Weight (FW) content, but also more salts and 275

trace elements. In all concentrate samples, concentrations of trace elements were however 276

below the Flemish legislation criteria [24] for use as fertilizer and/or soil conditioner in 277

agriculture. 278

Finally, waste waters produced by both acidic and alkaline air scrubbers were sampled 279

and physico-chemically analyzed (Table 5). 280

[Table 5 here] 281

Results show that the pH of the acidic waste water was continuously in the range of 2 to 3, 282

while the pH of the alkaline water was around 9. The EC and the salt content of the acidic and 283

the alkaline waste water were both high. The nitrogen content of the acidic waste water was 284

about three orders of magnitude higher compared to that of the alkaline waste water. 285

286

3.2 Economic and ecological analysis 287

Twenty-one different cultivation scenarios were economically and ecologically evaluated 288

(Table 6-7; Fig. 2). 289

12

[Table 6 here] 290

[Table 7 here] 291

[Fig. 2 here] 292

In scenario 3 to 5 and 16 to 21 the economic cost was higher than that of the reference 293

scenario. Hereby in scenario 3, 16 and 17 also the ecological impact was higher as compared 294

to the reference. Interestingly, all the other scenarios under study had a significantly lower 295

ecological and economic impact than the common practice. 296

297

4. Discussion 298

4.1 Fertilizer value 299

In nutrient-rich regions crude digestates cannot or only sparingly be returned to agricultural 300

land. Nevertheless, analytical results show that application of this product can be beneficial. 301

Advantages while using digestates in comparison to conventional animal manure are for 302

example the higher C/N ratio, 3.9 vs. 1.9, and the higher nitrogen uptake efficiency (NUE), 303

which is the relative amount of NH4-N compared to the total amount of nitrogen, 81 vs. 65-69 304

% (Table 3; [9]). The reason for this is that through anaerobic digestion organic nitrogen is 305

converted into NH4, which is directly available for the plant [16]. Also, during the anaerobic 306

digestion easily biodegradable organic compounds are converted into biogas, while complex 307

molecules such as lignin stay behind [16]. Hereby, the digestate keeps its soil enhancing 308

properties. Furthermore, it was observed that the extraction efficiency of macronutrients (P, 309

K, Na, Ca, Mg) using NH4OAc-EDTA at pH 4.65 as an extraction agent, is higher (up to 100 310

%) for digestate derivatives than for conventional pig manure. This translates into a higher 311

nutrient availability for plants. A final interesting observation was that the N/P ratio is about 312

four times higher for liquid fractions of digestates than for animal manure, 13 vs. 3.4, and 313

approximately five times higher than for digestates as such, 13 vs. 2.8, as most of the 314

13

phosphorus ends up in the thick fraction after separation. In light of phosphorus becoming 315

more restrictive in legislative frameworks for soil nutrient application rates [19], this nutrient 316

has become the limiting element in allowed dosage of organic fertilizers. In this perspective 317

the use of phosphorus reduced liquid fractions of digestate is highly interesting, because more 318

nitrogen can be applied to the soil for the same amount of phosphorus. When mixing digestate 319

(φ = 0.5) and its liquid fraction (φ = 0.5), the relative amount of N to P stays high, while also 320

the soil structure enhancing properties (Ca, Mg, OC) increase. All of these benefits make it an 321

interesting opportunity for agriculturists to treat their manure by anaerobic digestion and re-322

use the digestate and/or its derivatives on soil, either as base fertilizer and/or as substitute for 323

mineral fertilizers that are based on fossil resources. 324

In addition, results indicate that concentrates produced by membrane filtration have 325

potential as N-K fertilizer. The observed N content was 6.4±0.4 kg t-1 FW, comparable to 326

conventional pig manure, 5-10 kg t-1 FW [9]. The average NUE was 78 %, which is higher 327

than conventional pig manure, 69 % in this study or 65 % in [9]. Furthermore, waste water 328

from acidic air scrubbers shows potential as N-S fertilizer. The N content was 23±9 kg t-1 FW 329

and the S content was approximately 34 kg t-1 FW. Both the N and the S extraction efficiency 330

were 100 %, which is a prerequisite for recognition as a valuable mineral fertilizer according 331

to the EU requirements for ‘sulphate of ammonia’ (EC 2003/2003 [25]). Finally, waste water 332

from alkaline air scrubbers was poor in nutrients and therefore exhibits no potential as green 333

fertilizer. 334

From the above it can be concluded that the classification of digestate and its derivatives 335

should be reconsidered on national and European scale, with attention for their qualitative 336

fertilizer properties. The legal categorization of such derivatives as ‘green mineral fertilizers’ 337

next to the existing framework of ‘fossil-based mineral fertilizers’ (EC 2003/2003 [25]) might 338

be indispensable for their success in the European Union. 339

14

340

4.2 Potential bottlenecks for re-use 341

When using acidic air scrubber water in agriculture, one should be aware of some practical 342

limitations. At first, the low pH (2 to 3) of this waste stream shows that this waste water does 343

not only contain ammonium sulfate, but also significant amounts of sulfuric acid. As a 344

consequence, this product has acidifying and corrosive properties. It is therefore advised to 345

use corrosion-durable injectors and to avoid direct contact with skin and plants. Another, 346

more practical solution for this problem is to mix the acidic with the alkaline waste water (pH 347

9), thereby neutralizing the pH, or at least maintaining a weak acidic pH to avoid unwanted 348

ammonia emissions. Next, it is important to indicate that while mixing or storing the acidic 349

waste stream, H2S can be released which is very toxic even at low concentrations. This can be 350

explained by the presence of sulfate reducing bacteria, which are able to use sulfates for the 351

oxidation of organic compounds or hydrogen under low-oxygen conditions, a process in 352

which H2S is produced [26]. For all of these reasons, best practice requirements for 353

implementation and use of waste water from acidic air scrubbers should be set up in order to 354

minimize health risks and to prevent soil degradation. Finally, a critical point when using 355

acidic waste water in agriculture could be the salt content. Results show that the EC of this 356

stream is 112±42 mS cm-1, while that of conventional pig manure amounts to 35 mS cm-1, 357

which is in line with literature data, 30-50 mS cm-1 [9] and [27]. Too high salt contents can 358

cause soil salinification and can dramatically reduce crop production [28] and [29]. 359

As for acidic air scrubber water, results show that also membrane filtration concentrates 360

have elevated salt contents, 60±8 mS cm-1, resulting in high salt/N ratios (up to 6) for this 361

product. Therefore, when using concentrates in agriculture, it is important to pay attention to 362

the salt doses per unit nitrogen applied to the soil. Next, high sodium adsorption ratios (SAR), 363

which are ratios of monovalent cations such as K and Na to divalent bases such as Ca and Mg, 364

15

can cause soil structure degradation, especially when soils are rich in clay [28]. Finally, 365

results show that the K2O content of the concentrates produced by the 1st filtration was 366

5.2±3.2 kg t-1 FW, which is slightly lower than the expected range of [30], but still higher than 367

that of conventional pig manure, 4.3 kg t-1 FW. Although this element can be important for 368

crop production, high ratios of potassium to nitrogen are not preferred in every agricultural 369

application. Particularly livestock-farmers rather use potassium-poor fertilizers, because of the 370

potential health risks for cattle, such as head illness, at high potassium fertilization, > 50 t ha-1 371

y-1 [31] and [32]. Therefore, depending on the composition of the base fertilizer and the soil 372

characteristics, more or less concentrate can be applied as mineral fertilizer, with a maximum 373

advised K2O-dose of 70 kg ha-1 y-1 [31] and [32]. 374

None of the analyzed products exceeded the legal composition and use requirements of 375

heavy metals for re-use as fertilizer and/or soil conditioner in agriculture, as described in 376

Flemish legislation [24]. Only for one sample of dry thick digestate the amount of Ni slightly 377

exceeded the legal value. Furthermore, it should be remarked that because the relative amount 378

of NH4-N compared to the total amount of N is higher for digestates and its derivatives than 379

for conventional animal manure, emission poor application techniques, for example direct 380

injection, should be used. Also, fields must be ploughed as soon as possible after application 381

of these fertilizers in order to minimize NH3-emissions to air. Research on the microbiological 382

quality of digestate and its derivatives was not included in this study, but will be aspect of 383

future research. Yet, an orientating study [16] demonstrated that the amount of both aerobic 384

and plant pathogens in digestate is less than in animal manure, while the amount of anaerobic 385

pathogens is higher. More research in this field is, however, required. 386

387

4.3 Economic and ecological benefits 388

Re-use of valuable nutrients coming from digestate processing as a substitute for artificial 389

16

fertilizers could result in significant fossil energy and CO2-emission savings, as well as cost 390

savings. The energy consumption for artificial fertilizer use (N, P2O5, K2O) in the reference 391

scenario was 3.5 GJ ha-1, which results in a GHG-emission of 193 kg ha-1 y-1 in terms of CO2-392

equivalents, assuming that natural gas was used for the production of artificial fertilizers and 393

that diesel was used for the transport and application of fertilizers. The economic fertilizer 394

cost for the agriculturist in this scenario amounts to 54 € ha-1. It is observed that the 395

substitution of artificial fertilizers by acidic waste water results in significantly less economic 396

and ecological costs (scenario 1-2), because the impact of artificial fertilizer production 397

diminishes. This is also the case when artificial fertilizers are substituted by membrane 398

filtration concentrates (scenario 14-15). On the contrary, substitution of artificial fertilizers by 399

digestates (Scenario 16-17) and mixtures of digestate (φ = 0.5) and LF digestate (φ = 0.5) 400

(Scenario 20-21) requires more artificial N than the common practice, because of the low 401

N/P2O5 ratio of these products. Therefore, these scenarios seem not interesting. It should 402

however be remarked that in this study the relative amount of available N compared to the 403

total amount of N in the digestate derivatives was assumed to be 60 %, according to the 404

manure regulation. Nevertheless, results indicate that the actual amount of available N in 405

digestate derivatives is higher. The implementation of a new categorization for these products 406

could therefore result in a lower economic and ecological impact than predicted in the current 407

study. Furthermore, substituting artificial fertilizers with LF digestate turns out in 408

significantly less artificial K2O requirements, while no artificial N is necessary. This results in 409

a very low ecological impact: an energy use of 0.77 GJ ha-1 and a GHG-emission of 55 kg ha-1 410

expressed in terms of CO2-equivalents. However, because the N concentration in the liquid 411

fraction is low, the application and transportation costs for artificial fertilizers in this scenario 412

are higher than in the common practice. 413

Substituting animal manure with biomethanisation digestates as base fertilizer (Scenario 3-414

17

4-5) results in more artificial N requirements than the common practice, because the ratio of 415

N to P is lower for digestates than for animal manure. This results in a higher economic 416

impact than the common practice for all these scenarios and a higher ecological impact for 417

scenario 3. Also additional artificial P2O5 can be required in these scenarios. On the other 418

hand, substitution of animal manure with LF digestate (Scenario 6-7-8) results in significantly 419

less artificial K2O use. This results in economic and ecological benefits for the agriculturist. 420

However, also in these scenarios additional artificial P2O5 can be required. Substituting animal 421

manure by a mixture of digestate (φ = 0.5) and LF digestate (φ = 0.5) (Scenario 9-10-11) also 422

results in less artificial K2O use than the common practice, while maintaining a high N and 423

P2O5 dose. When simultaneously substituting artificial fertilizers with membrane filtration 424

concentrates, the highest economic benefits were reached, up to 82 € ha-1 (Scenario 13), while 425

a relative reduction of 65.4 % in the ecological impact was obtained compared to the common 426

practice. 427

Finally, it is interesting to notice that significant amounts of sulfur are applied to the soil 428

in scenario’s 1, 2, 4, 5, 7, 8, 10, and 11. This could result in an extra economic benefit of 0.75 429

€ kg-1 S (personal communication, Triferto, 2010), depending on the sulfur need of the 430

agricultural crops. 431

432

5. Conclusion and future perspectives 433

Recuperation and cradle-to-cradle re-use of macronutrients from digestate derivatives can be 434

an important aspect in the further development of sustainable agriculture, anaerobic digestion 435

and green chemistry. Concentrates following membrane filtration through reversed osmosis 436

show potential as green N-K fertilizer, whereas waste water from acidic air scrubbers shows 437

potential as green N-S fertilizer. Important bottlenecks for agricultural re-use of concentrates 438

could be the salt content, the sodium adsorption ratio and the potassium content, especially for 439

18

livestock-farmers. Bottlenecks for agricultural re-use of acidic air scrubber water could be the 440

pH, the salt content and its corrosive properties. Substituting artificial fertilizers by acidic air 441

scrubber water or membrane filtration concentrates theoretically always results in significant 442

economic and ecological benefits for the agriculturist. The highest economic benefits were 443

obtained when animal manure is substituted by a mixture of digestate (φ = 0.5) and liquid 444

fraction of digestate (φ = 0.5), while artificial fertilizers are replaced with membrane filtration 445

concentrates. Field research is now on-going in order to evaluate the impact on soil and crop 446

production by application of these new green fertilizers. 447

448

Acknowledgements 449

The authors would like to thank the Agency for Innovation by Science and Technology for 450

financing the IWT/KMO-project 90831. 451

452

453

454

455

456

457

458

459

460

461

462

463

464

19

465

466

467

468

469

References 470

[1] Fertilizers Europe. Forecast of food, farming and fertilizer use in the European Union 471

2010-2020. Brussels: European Fertilizer Manufacturers Association; 2010. 20 p. 472

[2] Fertilizers Europe. Understanding nitrogen and its use in agriculture. Brussels: European 473

Fertilizer Manufacturers Association; 2004. 64 p. 474

[3] Öborn I, Andrist-Rangel Y, Askekaard M, Grant C, Watson C, Edwards A. 475

Critical aspects of potassium management in agricultural systems. Soil Use Manage 476

2005;21(1):102-12. 477

[4] Ruddock J, Short TD, Brudenell K. Energy integration in ammonia production. In: 478

Brebbia CA, Sakellaris I, editors. 1st International Conference on Sustainable Energy, 479

Planning and Technology in Relationship to the Environment; 2003, May; Halkidiki, Greece. 480

Ashurst: Wit Press; 2003. p. 267-276. 481

[5] Smit AL, Bindraban PS, Schröder JJ, Conijn JG, Van Der Meer HG. Phosphorus in 482

agriculture: global resources, trends and developments. Wageningen: Plant Research 483

International BV; 2009, September. 36 p. Report No.: 1566-7790,282. 484

[6] Vilalba G, Liu Y, Schroder H, Ayres RU. Global phosphorus flows in the industrial 485

economy from a production perspective. J Ind Ecol 2008;12(4):557-69. 486

[7] Laherrere JH. Learn strengths & weaknesses to understand Hubbert curve. Oil Gas J 487

2000;98(16). 488

20

[8] Fehrenbach H, Giegrich J, Reinhardt G, Sayer U, Gretz M, Lanje K, et al. Criteria for a 489

sustainable use of bio-energy on a global scale. Dessau-Roßlau: German Federal 490

Environment Agency; 2008, January. 245 p. Report No.: UBA-FB20641112. 491

[9] Lemmens E, Ceulemans J, Elslander H, Vanassche S, Brauns E, Vrancken K. Best 492

available techniques (BAT) for animal manure processing. 3rd ed. Ghent: Academia Press; 493

2007. 337 p. 494

[10] Hjorth M, Christensen KV, Christensen ML, Sommer SG. Solid-liquid separation of 495

animal slurry in theory and practice. A review. Agron Sustain Dev 2010;30(1):153–80. 496

[11] Umetsu K, Kondo R, Tani M, Hayashi T. Fertilizer value of anaerobically co-digested 497

dairy manure and food processing wastes. In: Takahashi J, Young BA, editors. 1st 498

International Conference on Greenhouse Gases and Animal Agriculture; 2001, November 07-499

11; Obihiro, Japan. Amsterdam: Elsevier Science BV; 2002. 12 p. 500

[12] Masse L, Masse DI, Pellerin Y. The use of membranes for the treatment of manure: A 501

critical review. Biosyst Eng 2007;98(4):371-80. 502

[13] Vaneeckhaute C, Meers E, Michels E, Christiaens P, Tack FMG. Fate of macronutrients 503

in water treatment of digestate using vibrating reversed osmosis. Water Air Soil Poll 2011, 504

October 11; doi: 10.1007/s11270-011-0967-6. 505

[14] Bonmati A, Flotats X. Air stripping of ammonia from pig slurry: Characterization and 506

feasibility as a pre- or post-treatment to mesophilic anaerobic digestion. Waste Manage 507

2003;23(3):261-72. 508

[15] Braungart M, Mc Donough W. Cradle-to-cradle: Remaking the way we make things. US: 509

North Point Press; 2002. 193 p. 510

[16] Calus A, Dumoulin A, Ghekiere G, Mattheeuws B, Platteau W, Tavernier P, et al. 511

Anaerobic digestion on farm scale: Techniques, resources and end products. Rumbeke-512

21

Beitem: Provincial Research and Advise Centre for Agriculture and Horticulture; 2007. 32 p. 513

Report No.: D/2007/0248/27. 514

[17] Van Ranst E, Verloo M, Demeyer A, Pauwels JM. Manual for the soil chemistry and 515

fertility laboratory: analytical methods for soil and plants equipment, and management of 516

consumables. Ghent: University of Ghent, International Training Centre for Post-Graduate 517

Soil Scientists; 1999. 243 p. 518

[18] Weaver RW, Angle S, Borromley P, Bezdiecek D, Smith S, Tabatabai A, et al. Methods 519

of soil analysis, Part 2. Microbiological and biochemical properties. USA: Soil Science 520

Society of America; 1994. 1121 p. 521

[19] Decree concerning the protection of water against nitrate pollution from agricultural 522

sources. Decree No.: BS13.05.2011-MAP4 (May 13, 2011). 523

[20] Gellings CW, Parmenter KE. Energy efficiency in fertilizer production and use. In: 524

Gellings W, Kornelis B, editors. Efficient use and conservation of energy [Internet]. Oxford: 525

Eolss Publishers; 2004 [cited 2011 December 29]. 14 p. Available from: 526

http://www.eolss.net/Sample-Chapters/C08/E3-18-04-03.pdf. 527

[21] Department of Energy and Climate Change. Guidelines to Defra 2011: GHG conversion 528

factors for company reporting. London: Department for Environment, Food and Rural Affairs; 529

2011. 50 p. 530

[22] Wiens MJ, Entz MH, Wilson C, Ominski KH. Energy requirements for transport and 531

surface application of liquid pig manure in Manitoba, Canada. Agr Syst 2008;98(2):74-81. 532

[23] Fuel Prices [Internet]. Brussels: Europe’s Energy Portal; c2006-11 [updated 2011 533

October 31; cited 2011 October 31). Available from: http://www.energy.eu/. 534

[24] Flemish Regulation concerning Waste Prevention and Control (Vlarea). Decree No. 535

D/2005/5024/13 (July 2, 1989). 536

22

[25] Regulation (EC) No 2003/2003 of the European parliament and of the council relating to 537

fertilizers. Pub. Law EC No. 2003/2003 (October 13, 2003). 538

[26] Kool A, Timmerman M, De Boer H, Van Dooren HJ, Van Dun B, Tijmensen M. 539

Compilation of knowledge on co-digestion. Culemborg: CLM Research and Advise BV, P-540

ASG, Ecofys; 2005. 102 p. Report No.: 90-5634-196-0. 541

[27] Moral R, Perez-espinosa A, Moreno-caselles J, Paredes C, Rufete B. Salinity, organic 542

content, micronutrients and heavy metals in pig slurries from southeastern Spain. Waste 543

Manage 2008;28(2):367-71. 544

[28] Municipal Support Division, Office of Wastewater Management, & Technology Transfer 545

and Support Division, Office of Research and Development. Usepa. Guidelines for water 546

reuse. Washington DC: United States Environmental Protection Agency; 2004, September. 547

478 p. Report No.: EPA/625/R-04/108. 548

[29] Verlinden G. Valorisation of effluents from manure processing. Heverlee: Pedological 549

Service of Belgium; 2005, April. 329 p. Report No.: R004. 550

[30] Burton CH. The potential contribution of separation technologies to the management of 551

livestock manure. Livest Sci 2007;112(3):208-16. 552

[31] Hillel D. Soil in the environment. Crucible of terrestrial life. New York: Academic Press; 553

2008. 307 p. 554

[32] Romheld V, Kirkby EA. Research on potassium in agriculture: Needs and prospects. 555

Plant Soil 2010;335(1-2):155-80. 556