evaluation of environmental impacts of citric acid and ... · 18 uses citric acid and glycerol...

1

Evaluation of environmental impacts of citric acid and glycerol outdoor 1

softwood treatment: case-study. 2

Essoua Essoua Gatien Gerauda, Beauregard Roberta, Ben Amorb, Blanchet Pierrea, Landry Veronicc 3

4

Contact information: a:Chaire Industrielle de Recherche sur la Construction Ecoresponsable en Bois (CIRCERB), 5

Pavillon Gene-H-Kruger, Université Laval 2425, rue de la terrasse Québec (Québec) G1V0A6 ; b : Department of Civil 6

Engineering, Université de Sherbrooke, 2500, boul. de l’université, Sherbrooke, QC, J1K 2R1, Canada; c: 7

FPInnovations, 319 rue Franquet, Québec, QC, G1P 4R4, Canada. * Corresponding author: 8

[email protected] 9

10

Keywords: Life cycle assessment; Outdoor wood siding; Biobased wood treatment; Residential building 11

12

Abstract 13

Over the last few decades, wood modification has been performed to improve wood product technical 14 performance. Using renewable based chemicals for wood modification is an innovative alternative to the non-15 renewable petrochemicals commonly used. However, it should be kept in mind that having the raw material 16 from renewable sources does not guarantee zero environmental impacts. In this study, the treatment considered 17 uses citric acid and glycerol mixture; two chemical products derived from renewable sources. In the residential 18 building context of Quebec-Canada, the cradle-to-grave life cycle assessment for untreated and treated 19 lodgepole pine wood siding was performed and compared. The results obtained show that the treated wood 20 siding has higher environmental impacts than the untreated wood siding, in spite of its longer service life. This 21 is partially caused by the high contribution of citric acid production used for treatment. The current service life 22 expectancy of treated wood siding was estimated to be 2.8 times longer than the one of untreated wood siding 23 based on standardized durability test and classification (AWPA E 10-12 and ASTM D 2017-05). Sensitivity 24 analysis showed that life cycle impacts of treated wood siding become lower than those from untreated wood 25 siding when service life expectancy reaches 5-times that of untreated wood siding. Life cycle assessment could 26 be used for guidance in developing better treatments to improve their environmental impacts. 27

28

29

30

31

32

33

34

35

36

37

38

39

2

1. Introduction 40

1.1. Residential buildings in Québec 41

In Canada, the building industry is the most important industries. It represents more than 6 % of the 42 Canadian gross domestic product (GDP) (Statistics Canada, 2013). Regarding expenditure assess between 43 2004 and 2013, residential sector (construction and renovations) represents more than 45 % of total expenditure 44 in this industry (CCQ, 2014). The government of Quebec, in order to tackle Climate change and other 45 environmental issues, adopted a sustainable development policy. Sourced in a responsibly way, renewable 46 materials such as wood in buildings is an option to reduce the environmental footprint of the building industry 47 (Ramage et al., 2017). Residential and commercial buildings consume around 50 % of the resources extracted 48 and 33 % of all Canadian energy. This industry produces around 25 % of waste, 10 % of particles in the air 49 and 35 % of GHG (ISED, 2015). Again, the government of Quebec is committed to promote the use of wood 50 material in the building industry in order to reduce its environmental footprint (Béchard, 2008; MFFP, 2015). 51 Regarding the exterior layer of the building envelope, wood represents a good alternative to the common siding 52 materials. 53

1.2. Exterior wood siding 54

The use of wood material in building projects has recently enjoyed an increasing support from 55 architects, which has led to a global trend in favor of wood-based high-rise non-residential and multi-56 residential buildings (Zeitler and De Jager, 2014). One main reason for this trend is that wood products are 57 perceived as environmentally friendly (Puettmann et al., 2010; Frühwald, 2007). By comparison with other 58 building materials such as polymers, wood polymers, stone, cement, concrete, bricks, steel, alumina, etc., wood 59 offers significant environmental benefits (EPA, 2007; Salazar and Meil, 2009; Frühwald, 2007; Ross, 2010; 60 Glew et al., 2012; Kim and Song, 2014; Fionnuala et al., 2015). Some of these benefits are biogenic carbon 61 storage, biodegradability and recycling plus its appearance and insulation properties. 62

Wood siding used in outdoor applications present some disadvantages such as moisture sensitive, 63 dimensional stability and fungal decay (Essoua Essoua et al., 2016; Hill, 2006). Generally, to solve these 64 issues, chemical or heat treatment can be applied. In this study, the chemical treatment considered is composed 65 by citric acid and glycerol (CA-G) mixture, developed by Essoua Essoua et al. (2016). This treatment was 66 performed to improve the durability (i.e. life span) of outdoor softwood products. Using renewable chemical 67 products such as citric acid and raw glycerol represents one important point of this treatment, since it is 68 expected that the improved product could also contributes in reducing the final environmental footprint of the 69 building envelope. The hypothesis was confirmed or denied by LCA analysis. Gustafsson and Borjesson 70 (2007) and Tufvesson and Borjesson (2008) have showed that not all biobased chemical product are more 71 ecofriendly than those from petrochemical industry. Valorization of a biodiesel industry by-product is also a 72 good incentive as the quantity of produced raw glycerol by this industry increases year after year (Pagliaro and 73 Rossi, 2008; Yang et al., 2012; Anand and Saxena, 2011). In the wood modification activities, even if a green 74 chemistry approach is followed, the modification process requires the input of additional energy and 75 processing, which results in itself to additional environmental impacts. Therefore, does the improved durability 76 of the product offset the added footprint brought by the additional process, in the context of building’s envelope 77 application? To answer this question about the environmental profile of the wood siding product with or 78 without treatment, life cycle assessment (LCA) is the appropriate tool. 79

1.3 Environmental footprint of wood siding product 80

In the literature, there are some environmental studies of different siding materials. Some of them compared 81 the wood material to other materials used in residential buildings (Athena, 1998ab; Marceau and VanGreem, 82 2002). Wood siding production generally consumes less energy than other siding materials, from a life cycle 83 perspective (i.e. from cradle to gate) (Simard, 2009). The total energy (GJ/t) consumed to obtain one square 84 meter of clay brick, vinyl, wood and steel cladding material are 4.6; 2.7; 3.6 and 39.5 respectively (Athena, 85 1998ab; Athena, 1999a; Athena, 2002). Regarding the water consumption (l/t), wood siding consumes 39 times 86 less than vinyl and 150 times less than steel siding (Athena, 1999a; Athena, 2002). The potential effects of 87

3

these siding materials in the Climate change impact indicator were less in the case of wood material than for 88 the vinyl or steel materials (Athena, 2002; Athena, 1999a; Athena, 1998b). All these studies present wood 89 siding material as the best environmental choice in the building wall cladding in comparison to other cladding 90 materials (polymers, concrete, alumina, etc.). Ferreira et al. (2014) performed the LCA of maritime pine wood 91 board treated thermally and untreated. Their midpoint impact categories results show that cladding made by 92 thermally treated pine boards presents a better environmental performance than untreated ones. It is specifically 93 the case for Ozone depletion, Acidification, Eutrophication, Photochemical oxidation formation and Metal 94 depletion. Regarding Climate change, Human Health and Fossil depletion, untreated pine boards are better 95 than treated ones. 96 97

1.4 Modified wood siding 98

Modified wood shows improve technological performance than untreated (Essoua Essoua et al., 2016). 99 Hygroscopicity and durability are enhanced after modification (Essoua Essoua et al., 2016; Ferreira et al., 100 2014). Consequently, modified wood siding presents a longer service life expectancy than untreated (Menzies, 101 2013; Hill, 2006). Therefore, modified wood siding would be expected to have lower maintenance 102 requirements than untreated one also (Hill, 2006; Kattenbroek, 2007). 103

For the environmental aspects of the modified wood products, the treatment requires additional 104 product, process and energy. Modification processes, using chemicals, are associated to additional 105 environmental impacts. It is clear that any wood modification process will result in an increased environmental 106 impact, but using renewable or non-renewable treatment products will affect the magnitude of this additional 107 impacts (Hill, 2006; Cobut et al., 2015). However, environmental impacts will also depend on the quantity of 108 chemicals and on the processing required to render them suitable to be used for the modification reactions. 109 Most of the modification process use temperature higher than ambient and the choice of the energy source can 110 also greatly affect the environmental impact of the modified wood siding. 111

1.5 Research and scope 112

The main aim of this study is to establish what are the environmental impacts of treated Lodgepole 113 pine (Lp) wood siding. This is performed based on the longest service life expectancy (SLE) of treated siding 114 compared to the untreated Lp wood siding. Treatment improves wood properties, mainly dimensional stability 115 and durability properties. This contributes greatly to increase the SLE of outdoor wood products (Ferreira et 116 al., 2014; Hill, 2006). The various Damage impact categories were evaluated by LCA methodology. LCA from 117 cradle to grave of treated and untreated outdoor Lp wood siding, for a residential building was performed and 118 the environmental impacts were compared. Sensitivity analysis using different SLEs, alternative end of the life 119 scenario and alternative eco-design scenario (such as terephtalic acid) was performed in the case of treated 120 wood siding and compared with untreated scenario. 121

122

2. LCA Methodology 123

This paper methodology follows ISO 14040 series standards (2006). Simapro Software (version 8.1) was 124 used for the modelling. In addition, the Ecoinvent 3.0 database was also used for modelling (Swiss Center for 125 Life Cycle Inventories, 2013). The Ecoinvent 3.0 version includes some custom processes for the foreground 126 built on Ecoinvent 2.2. These were implemented manually in the version 3.0. Allocation procedure was used 127 in this study. Currently when by-product or co-product from manufacturing process were used, it has decided 128 to refer to the allocation factor from the Ecoinvent database or from the life cycle inventory study (Swiss 129 Center for Life Cycle Inventories, 2013). Finally, Impact 2002+ impact method was using to calculate the final 130 environmental impacts of treated versus untreated lodgepole pine wood siding. All these steps including their 131 appropriate references were detailed in the following sub-sections. 132

2.1 Product description 133

4

Wood siding studied in this research is a treated massive wood siding product compared to untreated 134 ones. Logs coming from the forest were sawn to produce rectangular section boards as shown in in Figure 1. 135 After drying, machining was performed to produce profiled siding. After profiling, the siding was paint by 136 different coating layers, depending of the coating used and the protection level, producer want to give to the 137 product. The machining step produce a wood waste used to produce the energy necessary to heat the plant in 138 the winter season and to dry wood lumber all year long. The siding market offers different profiles. The 139 differences between profiles come from architectural engineering design expectations. The used profile is one 140 of the most commonly seen in the market. 141

142

Fig. 1. Siding machining process 143

2.2 Goal and Scope: Functional Unit 144

The main function of outdoor wood siding is to cover the wall surface of residential buildings. The 145 considered functional unit is coving one square meter of a wall surface during a period of 20 years (Center for 146 clean Product, 2009; Ferreira, 2014; Marceau and VanGeem, 2002). Therefore, reference flows are defined by 147 the quantity of wood siding products (for both scenarios) necessary to fulfill the selected functional unit. The 148 assessed scenarios are presented in more details right below. 149

2.3 Goal and Scope: System boundaries 150

The Lodgepole pine wood siding product system is studied from cradle-to-grave. All life cycle stages 151 are considered, from the extraction of the raw material (round wood) to their end-of-life (once treated or 152 untreated). Comparative LCA is performed to compare treated and untreated Lp wood siding, considering the 153 additional impact of the treatment versus the additional life expectancy of the product, with or without the 154 maintenance stage. Figure 2 represents a simplified system boundary of the Lp wood siding product. 155

5

156

157

158

Fig. 2. Lodgepole pine siding system boundaries (for both treated and untreated). 159

2.4. Life cycle inventory (LCI) 160

Most of the LCI data was collected from an industrial partner producing solid wood siding. When no 161 data was available from the partner, especially for the wood treatment process, the unit processes were obtained 162 from the Ecoinvent 3.0 (including some custom processes for the foreground built on Ecoinvent 2.2, and 163 converted to version 3.0 format) tables 1-A and 1-B. Given that the project is realized in the jurisdiction of 164 Quebec, North America, some processes, such as the electricity grid and transport are representative or adapted 165 to this context. In Quebec, most of the electricity comes from hydropower and road transport is used more 166 often than railroad. 167

Table 1-A 168

Input data for the untreated and treated wood siding LCI for production life cycle stage 169

Life cycle stages Siding type Based on process Quantity Source of data Production Untreated coated

siding (1 m2) – pallets (0.000355 ton) – Waste is Wood chips for paper industry (0.000902 ton) and Sawdust (0.004808 ton)

Electricity, medium voltage, at grid/QC 2012 U AmN CIRAIG

2.2938 kWh Ecoinvent v 2.2

Transport, lorry >32t, EURO4/RER U AmN CIRAIG

0.645730 tkm Ecoinvent v 2.2

Transport, freight, rail, diesel/US U AmN CIRAIG


Timber system 0.026 m3 Ecoinvent v 2.2 Packaging film, LDPE, at plant/RER U AmN CIRAIG

0.08850 kg Ecoinvent v 2.2

VOC, volatile organic compounds


Treatment, sewage, to wastewater treatment, class 3/CH U AmN CIRAIG

0.002024 m3 Ecoinvent v 2.2

Treatment Citric acid {RER}| production | Alloc Def, U


Glycerine {CA-QC}| treatment of waste cooking oil,


6

purified, esterification | Alloc Def, U Hydrochloric acid, 36% in H2O, from reacting propylene and chlorine, at plant/RER U AmN CIRAIG


Purified terephtalic acid {CA-QC}| production | Alloc Def, U


Wood preservation, vacuum pressure method, outdoor use, ground contact | Alloc Def, U

3.2508 kg Ecoinvent v 3.0 modified for the chemicals products

Heat, softwood chips from forest, at furnace 300kW/CH U AmN CIRAIG

41.99904 MJ Ecoinvent v 3.0

Electricity, medium voltage, at grid/QC 2012 U AmN CIRAIG


170

Table 1-B 171

Input data for the untreated and treated wood siding LCI for others life cycle stages 172

Life cycle stages Siding type Based on process Quantity Source of data Distribution Treated (783,675

kg/m3) Transport, 53' dry van (Class 8) /AM U AmN CIRAIG 25t avec retour à vide


Transport, freight, lorry 16-32 metric ton, EURO5 {GLO}| market for | Alloc Def, U


Untreated (580,5 kg/ m3)

Transport, 53' dry van (Class 8) /AM U AmN CIRAIG 25t avec retour à vide


Installation Treated and untreated

Transport, freight, lorry 16-32 metric ton, EURO5 {GLO}| market for | Alloc Def, U


Maintenance Treated Electricity, low voltage, at grid/QC 2012 U AmN CIRAIG


Alkyd paint, white, 60% in H2O, at plant/RER U AmN CIRAIG


VOC, volatile organic compounds


End of the life Treated Recycling wood/RER U AmN CIRAIG


Transport, lorry 7.5-16t, EURO5/RER U CIRAIG Partage


7

Disposal, wood untreated, 20% water, to municipal incineration/CH U AmN CIRAIG


Untreated Recycling wood/RER U AmN CIRAIG


Transport, lorry 7.5-16t, EURO5/RER U CIRAIG Partage


Disposal, wood untreated, 20% water, to municipal incineration/CH U AmN CIRAIG


173

2.4.1. Raw material and transformation 174

The data of the first life cycle stage (cradle-to-gate) was collected mainly from an industrial partner 175 processing plan, using a questionnaire and live interviews with the partner’s staff. The data covers raw material 176 usually extracted by the partner company, and its characteristics, such as quantity, dimensions, provenance 177 and transport, to the siding product packaging, including sawing, drying, planing and internal transport 178 processes. Chemicals treatment products were found in Ecoinvent database (Swiss Centre for Life Cycle 179 Inventories, 2015). The latter includes chemicals production, acquisition and transportation processes. 180

2.4.2. Wood treatment modelling 181

Treated Lp wood siding modelling was based on processes identical to the untreated products, plus the 182 treatment step. In order to model the treatment, laboratory results published by Essoua Essoua et al., (2016) 183 were used in combination with the Ecoinvent 3.0 database. We modeled current impregnation methods from 184 the wood treatment industry. Input of the treatment process consists of citric acid, glycerol, water and 185 hydrochloric acid, electricity for equipment, including blending, autoclave impregnation system, drying step 186 after impregnation, esterification oven and an additional sanding step after cooling the esterified samples. The 187 studied esterification reaction for CA-G treatment produce an ester bond. Each ester bond emits one water 188 molecule. At the esterification temperature of 180 °C, the decomposition of citric acid emits a CO2 molecule. 189 After treatment there are some polymer layers around the sample surface. These were removed by the sanding 190 operation, producing sanding residues. All of these were taken into account as emissions and treatment residues 191 in our inventory sheet. 192

The treatment, based on the biodegradation laboratory results (Essoua Essoua et al., 2016), allows to 193 increase the durability class of the Lp wood siding, from durability class 4 (SLE ≤ 7 years) to durability class 194 1 (SLE ≥ 20 years). Considering 20 years as the SLE in our functional unit, treated siding life in service is 2.8 195 times longer than the untreated Lp wood siding. Consequently, to cover house walls for a period of 20 years 196 with untreated Lp siding, it was considered that 2.8 times the amount of treated sidings was required. 197

2.4.3 Distribution (Road transportation) 198

In Canada, more than 90 % of the manufacturing products transportation is performed by road trucking. 199 In the forest industry, truck transportation is also more commonly used than the rail. In this study, wood siding 200 distribution from the gate of the factory to the grave was modelled by using truck transportation. Different 201 transportation processes were used, including their corresponding distances and rate of loading, both 202 information provided by the industrial partner. The process from the gate to the wholesaler uses a 53 ft long 203 truck that has an average load of 17.56 tons, 25 tons being the maximum. Distance from the industry gate to 204 the wholesaler is 270 km. From the wholesaler to retailer uses, a lorry of 20 tons loaded at 50 %, and the 205 distance estimate is 100 km. From the retailer to the building site, the lorry of 20 tons was considered loaded 206 at 30 % and the distance estimate is 70 km. Regarding the end-of- life stage, the average distance from the 207

8

construction zone considered to the demolition waste treatment site is 16 km. A lorry of 16 tons was used and 208 load was considered to be at 70 %. All transportation unit processes come from the Ecoinvent database. 209

2.4.4 Installation and maintenance 210

Outdoor wood siding products are products generally sold with a warranty. They are installed above 211 ground and fixed with clips. The company provides installation instructions, including specifications on the 212 pneumatic stapler to be used for the fixation to keep for the warranty to remain valid. The siding is originally 213 provided, coated with three coats of paint, representing a total weight value of 0.373 kg per m2 of siding. In 214 the Canadian market, the warranted SLE without maintenance is between 3 to 8 years. This gap, depends on 215 the coating used: transparent, semi-opaque or opaque; the number of coats applied; and the chemical 216 composition of the coating. 217

In the life service of outdoor wood coating, one factor that contributes to the coating failure is wood 218 water content. The relative humidity (RH) of the environment where the siding wood product is installed varies 219 a lot. For every RH variation, the wood product takes or loses water, to reach a new equilibrium humidity (EH) 220 with its environment. This results in wood-dimensional changes, which in turn cause cracks in coating layers 221 (Williams et al., 2000; Williams, 2010). The reduction of wood dimensional variation, through wood 222 treatment, enhances coating durability (Williams, 2010; Evans et al., 2015). In Norway climate conditions, for 223 coated wood siding treated by acetic anhydride and 2 layers of opaque acrylic coating, the producer provides 224 a warranty to coating of 12 years compared to 8 years for untreated wood pine siding. In this case, the extended 225 SLE benefit from coating is 50 % compared to untreated wood siding. 226

In this study, based on the advice of our industry collaborator, the warranty on coating SLE was 227 considered to be 8 years for the treated siding. Based on this, every 8 years, customers have to apply 2 new 228 coats on the siding. This was modeled to be 0.249 kg of coating producing 0.0153 KgCOV/m2 of air emission. 229 Paintbrush impact was neglected. For untreated siding no coating maintenance was modeled because its SLE 230 is less than 8 years hence it was modeled as replaced instead of maintained. 231

2.4.5 End-of-life 232

In Canada, a large proportion of solid waste is eliminated in landfilling (Statistics Canada, 2012). (ACQ, 233 2016). In the jurisdiction of Quebec, in 2014, the government adopted regulations prohibiting wood waste 234 landfilling. With these regulations Quebec, requires that a second life be given to the wood coming from the 235 building sector. ACQ (2016) confirms that around 415 000 tons of wood waste landfilling has been recently 236 avoided in the province. Hence we modeled the he end-of-life scenario in this context for the Lodgepole pine 237 wood siding to be 100 % recycled. An alternative scenario with 100 % incineration has also been analysed. 238 The transportation at this stage was estimated from the building site to the waste processing center. For the 239 recycling scenario, discarded siding wood was modeled to be recycled to produce fiberboard, further 240 operations were attributed to fiberboard production, the cut-off method was used. For both untreated and 241 treated wood material, the recycling process considered was from the Ecoinvent database (Swiss Centre for 242 Life Cycle Inventories. 2015). 243

2.5 Life cycle impact assessment (LCIA) 244

LCIA can be performed by using various methods. Impact 2002+ method was chosen for impact 245 assessment because the aim of the study was to focus on Damage categories. This method has fifteen midpoint 246 categories and four endpoint damage categories, as shown in table 2. 247

Table 2 248

Impact 2002+ life cycle impact method and its categories (Adapted from Jolliet et al., 2010). 249

Impact assessment model Midpoint categories (problems)

Endpoint categories (damages)

Impact 2002+ Carcinogens Human Health (DALY)a

9

Non-carcinogens Respiratory inorganics Ionizing radiation Ozone layer depletion Respiratory organics Aquatic ecotoxicity Ecosystem Quality (PDF.m².yr)b Terrestrial ecotoxicity Terrestrial acid/nutria Land occupation Aquatic acidificationd Aquatic eutrophicationd

Global warming Climate Change (kg CO2 eq)

Non-renewable energy Resources (MJ primary)c Mineral extraction

250 a DALY: Disability-Adjusted loss of Life Years. This unit characterizes the disease severity, accounting for both mortality and morbidity. 251 252 b PDF.m².y: Potentially Disappeared Fraction of species over one m² during one year. This unit represents the fraction of species disappeared on 1m² 253 of earth surface during one year. 254 255 c MJ primary: Mega Joule primary. The unit measures the amount of energy extracted or needed to extract the resource. 256

d These indicators are available in midpoint categories only. 257

Before start with the results section, it’s necessary to present different scenarios short name used in various 258 graphics (table 3) 259 260 Table 3 261 262 Scenarios description 263

Scenarios Description Short name used Scenario 1 Untreated siding with recycling at 100% for the end of the life Untreated 20 y: base scenario with SLE

at 20 years, end of life 100 % recycling Scenario 2 Treated siding with citric acid and glycerol mixture. Recycling

at 100% for the end of the life Treated 20 y: base scenario with SLE at 20 years, end of life 100 % recycling

Scenario 1.1 SLE of untreated siding was estimate at 35 years Untreated 35 y: base scenario with SLE at 35 years

Scenario 2.1 SLE of treated siding was estimate at 35 years Treated 35 y: scenario base with SLE at 35 years

Scenario 1.2 SLE of untreated siding was estimate at 70 years Untreated 70 y: scenario base with SLE at 70 years

Scenario 2.2 SLE of treated siding was estimate at 70 years Treated 70 y: scenario base with SLE at 70 years

Scenario 1.3 Untreated siding with incineration at 100% for the end of the life stage

Untreated incineration: base scenario with SLE at 20 years, end of life 100 % incineration

Scenario 2.3 Treated siding with incineration at 100% for the end of the life stage

Treated incineration: base scenario with SLE at 20 years, end of life 100 % incineration

Scenario 2.4 Treated siding with terephtalic acid and glycerol mixture. Recycling at 100% for the end of the life

Alternative treatment terephtalic acid 20 y

264

3. Results 265

3.1 Comparative results and interpretation 266

This comparative assessment results between untreated and treated Lp wood siding are presented and 267 discussed below. In coherence with Table 1, the results are both presented at the midpoint and endpoint level 268 (see Figs 3 and 4). As a reminder, two scenarios are shown: Scenarios 1 and 2 refer to untreated and treated 269

10

Lp wood siding, respectively. For the same life in service of 20 years, scenario 1 indicates the use of 2.8 time 270 untreated wood siding while scenario 2 indicates the use of one treated wood siding. 271

For the midpoint categories, Figure 3 shows that in most categories, treated siding presents more 272 environmental impact than the untreated one. Of interest the global warming and non renewable energy 273 consumption are in the order of 45% to 50% for untreated siding when compared to treated siding. Treated 274 siding presented 44% less environmental impacts only for land occupation and 80% less impact for respiratory 275 organics. Needing 2.8 times more untreated siding to fulfill the functional unit over 20 years, the required land 276 surface to produce the necessary quantity of wood becomes important. 277

11

278

279

280

Fig. 3. LCA comparison between untreated and treated: Midpoint categories 281

12

Aggregation results of the Mid-point to have End-point categories as shows in table 1, are presented 282 in Figure 4. The latter shows that for the human health damage categories, the Respiratory organics do not 283 carry enough weight to change the overall result (i.e. treated siding presents more environmental impact than 284 the untreated one). This is not the case for the Ecosystem quality damage categories, as previous observations 285 with land occupation midpoint categories remains. The Land occupation has a higher damage factor in the 286 Ecosystem quality endpoint categories (Jolliet et al., 2010). Finally, untreated Lp wood siding produces 35 %, 287 54 % and 51 % less impact than the treated siding referring to Human health, Climate change and Resources 288 damage categories, whereas only for the Ecosystem quality damage, treated Lp wood siding shows a better 289 environmental profile than the untreated one (difference of 50 %). To supporting these information, absolute 290 values are given in table form, for each damage category in the Figure 4. 291

292

Dommages categories Unit Untreated 20 y Treated 20 y Human health DALY 6,5674167E-6 1,0645991E-5

Climate change kg CO2 eq 4,0802217 9,6408761 Resources MJ primary 73,796384 169,71902

Ecosystem quality PDF*m2*yr 40,731482 20,413851

293

Fig. 4. LCA comparison between untreated and treated: Damages categories and absolute values of impacts 294 categories in table form. 295

296

3.2 Contribution analysis results and interpretation 297

Life cycle of wood siding treated were simplified in five stages: Production, distribution, installation, 298 maintenance and end of the life. As explain before, production stage includes what said in section 2.4.1 and 299 section 2.4.2. The life cycle contribution analysis of treated Lp wood siding was performed to highlight the 300 life cycle stage with a significant contribution (figure 5). The production stage presents the most environmental 301 impact in all endpoint categories, followed by the maintenance stage. As supporting information, absolute 302 values are shows in table form in figure 5. 303

Damages categoriesHuman health Ecosystem quality Climate change Resources

Tot

al im

pact

(%

)

100

90

80

70

60

50

40

30

20

10

0

Untreated 20 y Treated 20 y

100 %

49 % 46 %

100 %

50 %

100 % 100 %

65 %

13

304

Dommages categories

Unit Total Production Distribution Installation Maintenance End of the life

Human health DALY 1,0645991E-5 9,4890189E-6 3,2934006E-10 6,9125341E-10 1,1559402E-6 1,1449769E-11 Ecosystem

quality PDF*m2*yr 20,413851 19,750766 0,0002754692 0,0021510725 0,66065197 6,2790176E-6

Climate change kg CO2 eq 9,6408761 7,9857372 0,00035620629 0,00045394652 1,6543076 2,1148229E-5 Resources MJ primary 169,71902 135,96129 0,0058180772 0,012382868 33,739172 0,00035633857

305

Fig. 5. Life cycle contribution analysis using endpoint impact categories (scenario treated Lp wood siding) 306 and absolute values of impacts categories in table form. 307

308

Within the production stage, the highest impact came from the treatment step. The treatment step consumes 309 a significant amount of energy and material. The Sankey diagram information of treated and untreated coated 310 siding is presented in table 4. Within the LCA of treated coated siding, referring to the resources impact 311 category, the production stage contribution is 88.2 % in comparison to 11.8 % for the maintenance stage. 312 Moreover, as shown in figure 6, it becomes obvious that the treatment step is an important hotspot (54.5 %). 313 This is mainly caused by citric acid production (38.6 %). 314

In the case of untreated coated siding, table 4 shows that production stage is a responsible for more 315 than 99.9 % of the total resources impact in comparison to the other life cycle stages (distribution and 316 installation, end of life, etc.). Within the production stage, the main contribution comes from the timber system 317 process (extraction, sawing, drying and planning), with a total impact of 82.9 %. For the end of the life 318 scenarios, there are no energy and material substitution. 319

Table 4. 320

Life cycle stages and processes steps of cradle to grave LCA of treated and untreated wood siding (resources 321 damage category) 322

Analyses Life cycle stages Processes Under processes LCA (Cradle to grave) of treated siding coated (1P) 100%

Production (1P) 88,2%

Untreated coated siding production (1P) 33,7%

Timber system (0,0128 m3) 27,9%

Citric acid and glycerol treatment (1P) 54,5%

Citric acid (RER)|production| Alloc, Def, U (1,95 kg) 38,6 %

Maintenance (2P) 11,8%

Alkyd paint, white, 60% in H2O, at plant/RER U AmN (0,622 kg) 11,8%


Tot

al im

pact

s (%

)

100

90

80

70

60

50

40

30

20

10

0

Production Distribution Installation Maintenance End of the life

14

LCA (Cradle to grave) of untreated siding coated (1P) 100%

Production

Light fuel oil, burned in industrial furnace 1MW. (5,33 MJ) 2,36%

Transport, freight, rail, diesel/US U Am N CIRAIG (10,9 tkm) 5,08%

Timber system (0,0358 m3) 82,9%

Acrylic binder, 34% in H2O, at plant RER U Am N CIRAIG (0,212 kg) 2,47%

Packaging film, LDPE, at plant RER U Am N CIRAIG (0,122 kg) 2,78%

323

324

4. Sensitivity analysis 325

Until now, presented results show that the treated siding is less ecofriendly than the untreated siding. This 326 for Human health, Climate change and Resources damage categories, and that even if the treatment improved 327 the siding life expectancy by 2.8 times over the untreated one. Referring to the Canadian Wood Council (2016), 328 for the same wood species, usual chemical wood treatment is expected to bring average SLE between 5 and 329 10 times more than the untreated wood. Assuming that the assessed treatment in the present work could reach 330 such SLE, the effect of longer service time was modelled in order to assess the sensitivity of the conclusions. 331 For a longer service life expectancy of 35 years (in comparison to the base case scenario of 20 years), 3 332 maintenance paint coatings were modelled for the treated siding product (Scenario 2.1). This corresponds to 5 333 times the SLE of the untreated siding (Scenario 1.1). In the case of 70 years SLE, 7 maintenance paint coatings 334 were modelled (Scenario 2.2). This corresponds to 10 times the SLE of untreated siding (Scenario 1.2). 335

With the SLE at 35 years (scenario 2.1), treated Lp wood siding is more ecofriendly than untreated Lp 336 wood siding (scenario 1.1) for the Human health (6 % less impact) and Ecosystem quality (72 % less impact) 337 categories. However, for the Climate change and Resources damage categories, treated siding is less 338 ecofriendly, with respectively 30 % and 28 % more impact than untreated Lp wood siding (Fig 6). In this SLE, 339 maintenance stage increases the environmental impact of treated siding from 11.8 % to 15.4 %. For the same 340 life time, maintenance stage requires 0.954 kg of paint with 0.0459 kgCOV/m2 emission in the air, soil and 341 water. The table present in figure 6 and 7, gives absolute values of each damage category, to supporting these 342 results. 343

344


Tot

al im

pact

(%

)

100

90

80

70

60

50

40

30

20

10

0

Untreated 35 y treated 35 y

100 %

72 %

100 % 100 %

28 %

100 %

70 % 94 %

15


Ecosystem quality PDF*m2*yr 72,734789 20,744177 Climate change kg CO2 eq 7,2861101 10,46803

Resources MJ primary 131,77926 186,5886

345

Fig. 6. LCA comparison between untreated and treated, with SLE at 35 y: Damages categories and absolute 346 values of impacts categories in table form. 347

When we increase the SLE of the treated Lp wood siding to 70 years (scenario 2.2), the results change 348 dramatically. Indeed, environmental impacts of treated siding are much lower for all impacts categories than 349 the untreated siding (scenario 1.2). The environment impacts were respectively 42 %, 85 %, 8 % and 7 % less 350 than those of the untreated siding for Human health, Ecosystem quality, Climate change and Resources (Fig. 351 7). With the SLE increasing, environmental impacts of maintenance stage increases. For the SLE at 70 years, 352 the contribution of the maintenance stage in the global environmental impacts of treated siding is 29.9 %. Paint 353 used is 2.226 kg with 0.1071 kgCOV/m2 released in the air, soil and water. With increasing SLE of treated 354 wood siding, maintenance is the unique stage in the life cycle stages that increases in the environmental impact. 355

356




357

Fig. 7. LCA comparison between untreated and treated, with SLE at 70 y: Damages categories and absolute 358 value of impacts categories in table form. 359

Comparison between untreated and treated wood siding at different SLE and the difference between their 360 impacts values permit to define the follows figure. When the difference value is positive, treated wood siding 361 present lower environment impact than the untreated one while when negative, treated wood siding presents a 362 higher environment impact than the untreated one. Line zero in the graphic, indicates where untreated and 363 treated wood siding have the same environmental impacts values. Looking at figure 8 it can be observed that 364 when the SLE of treated siding reaches 55 years (7,9 times scenario 1 of untreated siding), all Endpoint impact 365 categories become favourable to the treatment scenario. At that SLE, it becomes environmentally interesting 366 to treat the Lp wood siding with the citric acid and glycerol mixture. 367

368


Tot

al im

pact

(%

)

100

90

80

70

60

50

40

30

20

10

0

Untreated 70 y Treated 70 y

100 % 100 % 100 % 100 %

58 %

15 %

92 % 93 %

16

369

Fig. 8. Comparison of environmental impact between untreated vs treated Lp wood siding at different Service 370 life expectancies 371

The alternative scenario regards the End-of-life of the siding (treated and untreated). The considered End-372 of-life scenario was recycling at 100 %. In Quebec, since 2015, the government prohibits the wood product 373 landfill. In this way, incineration scenario was also taken into account. It was performed at 100 % for both 374 siding products. Two scenarios were defined. For untreated siding, scenario 1.3 and for treated siding, scenario 375 2.3. In these scenarios, there are no energy and material substitution. The comparison between scenario 1.3 376 and scenario 2.3 is presented in figure 9. The environmental impact comparison between untreated and treated 377 is similar to the one shows in the previous comparison (figure 4), except the Human health categories. In this 378 damage category and as shows by absolute values in table form in the figure 9, treated siding is ecofriendly 379 than untreated one. 380

381

Dommages categories Unit Untreated incineration Treated incineration Human health DALY 1,3695116E-5 1,3444119E-5



382

‐40%

‐20%

0%

20%

40%

60%

80%

20 years 35 years 55 years 70 years

Residual environmen

t im

pact (%

)

Service life expectancy (years)

Human health (DALY)

Ecosystem quality (PDF*m2*yr)

Climat change (Kg CO2 eq)

Resources (MJ Primary)


Tot

al im

pact

(%

)

100

90

80

70

60

50

40

30

20

10

0

Utreated incinaration Treated incinaration

49 %

100 % 100 % 98 % 100 %

49 % 51 %

100 %

17

Fig. 9. LCA comparison between untreated and treated, with incineration as end of the life: Damages 383 categories and absolute values of impacts categories in table form. 384

In order to give more information about end of the life scenarios of treated siding, comparison between 385 recycling and incineration scenarios were performed. Graphic and absolute values of figure 10, presents the 386 results. Recycling scenario is less ecofriendly than incineration one. The ecosystem quality, climate change 387 and resources damages categories present similar results. The main reason regarding energy used. In the case 388 of incineration scenario, energy produces is generally re-used in the same system or for the building heating. 389 Using energy produced, reduces the incineration scenario impacts. For the human health damage category, 390 recycling scenario is ecofriendly than incineration. The incineration scenario presents high impact for 391 carcinogens and non-carcinogens mid point categories than recycling scenario. 392

393

394

395

Dommages categories Unit Treated 20 y recycling Treated 20 y incineration Human health DALY 1,0645991E-5 1,3444119E-5


Resources MJ primary 169,71902 154,9248 396

Fig. 10. LCA comparison between recycling and incineration scenarios as end of the life of the treated Lp 397 siding at 20 y of SLE. Absolute values of impacts categories in table form. 398

5. Alternative eco-design scenario 399

To improve the treated wood siding life cycle profile, various eco-design scenarios are proposed. The 400 use of citric acid to prepare treatment solution comes from renewable resources. However, its production is 401 deriving from microorganisms’ fermentation of sugar. It’s a bio based process. It is a long process that needs 402 energy and resources. Previous studies have shown that bio-based chemicals aren't necessarily more 403 environmentally preferable compared to industrial chemistry one (Foulet et al. 2015). Different carboxylic 404 acids can be used to explore industrial chemistry alternative to our base case treatment scenario, such as 405 terephtalic acid. The main difference with the citric acid is at production process level (industrial synthesis). It 406 came from synthesis of acetic acid and xylene. Production processes of both acids were taken from the 407 Ecoinvent database (Swiss Centre for Life Cycle Inventories. 2015). The new terephtalic acid glycerol mixture 408

0,00E+00

2,00E+01

4,00E+01

6,00E+01

8,00E+01

1,00E+02

1,20E+02

1,40E+02

1,60E+02

1,80E+02

DALY PDF*m2*yr kg CO2 eq MJ primary

Human health Ecosystem quality Climate change Resources

Absolute values

Damages categories

Treated 20 y recycling Treated 20 y incineration

18

is modelled as a new scenario (scenario 2.4). The LCA results comparing the new treated siding (scenario 2.4) 409 and the untreated one (scenario 1) are presented in figure 11. 410

Changing citric acid to terephtalic acid does not change the environmental impact observations when 411 comparing untreated to the treated siding. Tendency is similar to the one shown in figure 4. However, it 412 improves by 7 %, the environmental impact of treated siding for Ecosystem quality damage category compared 413 to untreated siding (scenario 1). Production processes comparison of both acids presents terephtalic acid as an 414 ecofriendly alternative to citric acid for the Climate change and Ecosystem quality impact categories figure 415 11. For the Ecosystem quality damage category, aquatic ecotoxicity damage values is 58.2 kg TEG water 416 compared to 334 kg TEG water for the citric acid. This explains why terephtalic acid improves the Ecosystem 417 quality damage categories. For the non-renewable energy midpoint impact category, terephtalic acid caused 418 more impact than citric acid (52.3 MJ primary compared to 48.5 MJ primary). Supporting by absolute values 419 in table form presented in figure 12, the eco-design scenario show the same environmental impact observations 420 as those discussed in the base case scenarios. 421

422

423

Damages categories Units Terephtalic acid Citric acid Human health DALY 4,7059736E-6 2,0281432E-6



424

Fig. 11. LCA comparison of production stage between terephtalic acid and citric acid. Absolute values of 425 impacts categories in table form. 426

427

0,00E+00

1,00E+01

2,00E+01

3,00E+01

4,00E+01

5,00E+01

6,00E+01

DALY PDF*m2*yr kg CO2 eq MJ primary

Human health Ecosystem quality Climate change Resources

Absolute values

Damages categories

Terephtalic acid Citric acid

19

428

Dommages categories Unit Untreated 20 y Alternative treatment terephtalic acid 20 y Human health DALY 6,5674167E-6 1,5291076E-5



429

Fig. 12. LCA comparison between untreated 20 y and treated with terephtalic acid in the solution 20 y: 430 Damages categories and absolute values of impacts categories in table form. 431

6. Conclusions 432

Performed chemical wood modification to enhance physical and chemical wood properties and 433 consequently to extend its service life expectancy is a good way to make wood siding products more 434 competitive than other cladding material. Even when the used chemical products come from renewable 435 sources, their environmental impacts need to be considered. In this study, a cradle to grave LCA of outdoor 436 Lodgepole pine wood siding was performed. A citric acid and glycerol treatment was applied to improve their 437 technical performance. 438

Comparative LCA between untreated and treated Lodgepole pine wood siding shows that treated siding 439 results to overall more environmental impact, which is mainly explained by the treatment stage. In this stage, 440 citric acid production is the most important hot spot followed by the drying step after impregnation. In the 441 LCA stages of both siding products (untreated and treated), production and maintenance stages are those with 442 the largest impact. Distribution, installation and end-of the life stages present the lowest impact. Sensitivity 443 analysis showed that if the treatment efficiency enabled actual service life expectancy of 55 years or more, the 444 treated siding would become more environmentally friendly than the untreated wood option. 445

An alternative eco-design scenario was assessed to try to decrease the treatment additional impact: to 446 replace citric acid by terephtalic acid. This alternative scenario did not change the environmental impact 447 conclusions. In all damage categories except Ecosystem quality, treated siding results in more impact 448 compared to untreated siding, even if the latter was replaced more often 2.8 times. 449

As future work, it would be of interest to perform a life cycle costing, to complete this study and to offer 450 to the siding industry a complete view of this treatment. Also future development of low environmental impact 451 treatments to improve the performance of wood siding is still an important research challenge. 452

453

Acknowledgements 454

The authors would like to acknowledge the National Science and Engineering Research Council of 455 Canada and FPInnovations for funding this research (RDCPJ 386935). We also acknowledge, the CIRAIG at 456 École Polytechnique de Montreal, for their contribution on the data modeling. 457


Tot

al im

pact

(%

)

100

90

80

70

60

50

40

30

20

10

0

Untreated 20 y Alternative treatment terephtalic acid 20 y

100 %

70 %

47 % 43 %

100 %

43 %

100 % 100 %

20

References 458

ACQ, 2016. Résidus de bois au Québec, Nouvelle réglementation et fin de l’enfouissement. Association des 459 constructeurs de Québec (ACQ). http://www.acqconstruire.com/dossiers/gestion-residus-construction-460 renovation-demolition/residus-de-bois-au-quebec-nouvelle-reglementation-et-fin-de-enfouissement.html 461

Anand, P., Saxena, R. K. 2011. A comparative study of solvent-assisted pretreatment of biodiesel 462

derived crude glycerol on growth and 1,3-propanediol production from Citrobacter freundi. New 463

Biotechnol. 29(2), 199-205 464

Athena, 1993. Raw materials balances, energy profiles and environmental unit factor estimates: structural 465 wood product. Athena Sustainable Materials Institute. Ottawa, Athena Institute. p. 36. 466

Athena, 1998a. Life cycle analysis of brick and mortar products. Athena Sustainable Materials Institute. 467 Ottawa, Athena Institute. p. 38. 468

Athena, 1998b. A life cycle analysis of solid wood and steel cladding. Athena Sustainable Materials Institute. 469 Ottawa, Athena Institute. p. 22. 470

Athena, 1999a. Life cycle assessment analyses of building envelope materials. Update and expansion. Athena 471 Sustainable Materials Institute. Ottawa, Athena Institute. p. 38. 472

Athena, 2002. Cradle-to-gate life cycle inventory: Canada and US steel production by mill type. Athena 473 Sustainable Materials Institute. Ottawa, Athena Institute. p. 171. 474

Béchard, C. 2008. Stratégie d'utilisation du bois dans la construction au Québec (Wood use strategy for 475 construction in the province of Québec). 20 pp. 476

CCQ, 2014. Statistiques annuelles de l’industrie de la construction 2013. Commission de la Construction du 477 Québec, direction de la recherche et de la documentation. pp. 1-62. 478

Center for clean products, 2009. Life-cycle assestment of cladding products: A comparison of aluminum, brick, 479 granite, limestone and precast concrete. Final report, University of Tennessee, US. 480

Cobut, A., Blanchet, P., Beauregard, R., 2015. The environmental footprint of interior wood doors in non 481 residential buildings-part 1: life cycle assessment. Journal of Cleaner Production. (2015), 1-15 482

Canadian Wood Council, 2016. Treated wood – Preservative treated wood – advantage. Canadian Wood 483 Council. http://cwc.ca/wood-products/treated-wood/preservative-treated-wood/advantages/. Consulted 01- 484 2016 485

EPA, 2007. Energy Trends in Selected Manufacturing Sectors: Opportunities and Challenges for 486 Environmentally Preferable Energy Outcomes. ICF International. http://www.epa.gov/sectors/pdf/energy/ch3-487 5.pdf. Consulted in February 2015. 488

Essoua E. G. G., Blanchet, P., Landry, V., Beauregard, R., 2016. Pine wood treated with a citric acid and 489 glycerol mixture: Biomaterial performance improved by a bio-byproduct. Bioresources.com. 11, 3049-3072. 490

Evans P. D., Haase J. G., Shakri A., Seman B. M., Kiguchi M., 2015. The search for durable exterior clear 491 coating ofr wood. Coating. 830-864. 492

Ferreira, J., Bruno Esteves, B., Nunes, L., Domingos, I., 2014. Life cycle assessment of thermally treated and 493 untreated maritime pine boards: A Portuguese case study. European Conference on Wood Modification. 1-8. 494

Fionnuala, M., Ger, D., Kevin, M., 2015. Greenhouse gas and energy based life cycle analysis of products 495 from the Irish wood processing industry. Journal of cleaner production. 92, 134-141. 496

Foulet, A., Birot, M., Sonnemann, G., Deleuze, H., 2015. Life cycle assessment of producing emulsion-497 templated porous materials from Kraft black liquor - Comparison of a vegetable oil and a petrochemical 498 solvent. Journal of cleaner production. 91, 180-186 499

Frühwald, A., 2007. The Ecology of Timber Utilization Life Cycle Assessment Carbon Management 500 etc. Paper presented at the Probos Foundation, Doorn, Germany. 501 http://www.probos.nl/kennisnetwerk/pdf/PresentatieMVO20071115ArnoFruhwald.pdf. Consulted in may 502 2015. 503

21

Glew, D., Stringer, L.C., Acquaye, A.A., McQeen-Mason, S., 2012. How do end of life scenarios influence 504 the environmental impact of product supply chains? Comparing biomaterial and petrochemical products. 505 Journal of cleaner production. 29-30, 122-131. 506

Gustafsson, L., Börjesson, P., 2007. Life cycle assessment in green chemistry. A comparison of various 507 industrial wood surface coatings. Int J LCA 12(3):151-159. 508

Hill, C. A. S. 2006. Wood Modification: Chemical, Thermal and Other Processes, John Wiley & Sons Ltd, 509 Chichester, UK. DOI: 10.1002/0470021748 510

ISED, 2015. Feuille de route pour la durabilité des entreprises. Innovation, Sciences and Economic 511 Dévéloppement Canada. https://www.ic.gc.ca/eic/site/csr‐rse.nsf/eng/rs00585.html. Consulted 10-2015 512

ISO, 2006b. ISO 14040 Environmental management - Life cycle assessment - Principle and framework. 513 International Organisation for Standardisation (ISO). Geneva, Switzerland. p. 20. 514

ISO, 2006c. ISO 14044: Environmental management - Life cycle assessment - Requirements and guidelines 515 International Organisation for Standardisation (ISO). Geneva, Switzerland. p. 46. 516

Jolliet, O., Saadé, M., Crettaz, P. 2010. Analyse de cycle de vie : Comprendre et réaliser un écobilan. PPUR 517 Presses polytechniques. p. 302 518

Kattenbroek, B. 2007. The commercialisation of wood acetylation technology on a large scale. In: Proceedings 519 of the third European conference on wood modification. Cardiff, UK, pp. 19-22. 520

Kim, M.H., Song H.B., 2014. Analysis of the global warming potential for wood waste recycling systems. 521 Journal of cleaner production. 69, 199-207. 522

Marceau, M.L., VanGeem M. G., 2002. Life cycle assestment of a concrete masonry house compared to a 523 wood frame house. Portland Cemment Association report. 1-168. 524

Menzies, G., 2013. Service life plannng analysis of Accoys timber used in windows. Institut for building and 525 urban design, Heriot Watt University. 1-11. 526

MFFP, 2015. Le bois: matière première du développement responsable. Ministère de la foret, de la faune et 527 des parcs du Québec. p. 6. https://www.mffp.gouv.qc.ca/publications/forets/entreprises/charte-bois.pdf. 528 Consulted in 10/2015 529

Pagliaro, M., and Rossi, M. 2008. The future of glycerol: New uses of a versatile raw material. R. Soc. Chem. 530 117 531

Puettmann M.E., Bergman R., Hubbard S., Johnson L., Lippke B., Wagner F., 2010. Cradle to gate life 532 cycles invetories of U.S. wood products production – CORRIM Phase I and Phase II Products. Wood and 533 Fiber J., 42(CORRIM Special Issue). 15-28. 534 535 Ramagea, M. H., Burridgeb, H., Busse-Wicherc, M., Feredaya, G., Reynoldsa, T., Shaha, D. U., Wud, G., 536 Yuc, L., Fleminga, P., Densley-Tingleye,D., Allwoode, J., Dupreec, P., Lindenb, P.F., Scherman, O. 2017 537 The wood from the trees: The use of timber in construction. Renewable and Sustainable Energy Reviews 68 538 (2017) 333–359. 539

Ross, R.J., 2010. Wood handbook: wood as an engineering material. GTR-190, USDA Forest Service. 540

Salazar, J., Meil, J., 2009. Prospects for carbon-neutral housing: the influence of greater wood use on the 541 carbon footprint of a single-family residence. Journal of cleaner production. 17, 1563-1571. 542

Simard, E., 2009. Les matériaux de construction résidentielle dans une perspective durable : Analyse 543 comparative. Centre universitaire de formation en environnement : Université de Sherbrooke. pp. 179. 544

Statistics Canada, 2013. Canadian Building Industry. Chapter 6, pp. 1-10. http://www.statcan.gc.ca/pub/11-545 402-x/2011000/pdf/construction-eng.pdf 546

Statistic Canada, 2012. Section 3: Solid waste. http://www.statcan.gc.ca/pub/16-201-x/2012000/part-partie3-547 eng.htm 548

22

Swiss Centre for Life Cycle Inventories. 2013. Ecoinvent database V 2.2. http://www.ecoinvent.org/database/. 549 2012 550

Swiss Centre for Life Cycle Inventories. 2015. Ecoinvent database V 3.0. http://www.ecoinvent.org/database/. 551 2015 552

Tufvesson, L., Börjesson, P., 2008. Wax production from renewable feedstock using biocatalysts instead of 553 using fossil feedstock and conventional methods. Int J LCA (13):328-338. 554

Williams, S. R., 2010. Wood handbook, chapter 16: Finishing of wood. General technical report FPL-GTR-555 190. Madison, WI: US department of agriculture, forest service, forest products. pp. 1-37. 556

Williams, S. R., Jourdain, C., Daisey, G. I., Springate, R. W., 2000. Wood properties affecting finish service 557 life. Forest products laboratory, Madison, WI. US. 902, 35-42. 558

Yang, F., Hanna, M. A., Sun, R. 2012. Value-added uses for crude glycerol-A byproduct of biodiesel 559 production. Biotechnology and Biofuels 5, 13. 560

Zeitler Fletcher, S., De Jager, C., 2014. Summary report: Survey of international tall wood buildings. Report 561 for Forestry Innovation Investment (BC) and Binational Softwood Lumber Council (Canada-USA). p. 37. 562

evaluation of environmental impacts of citric acid and ... · 18 uses citric acid and glycerol...

Documents