079_D 3.3-3.5 Life-Cycle Assessment and Environmental Assessment

Proposal full title:

Algae and aquatic biomass for a sustainable production of 2nd generation biofuels

Proposal acronym:

AquaFUELs

Type of funding scheme:

Cooperation

Theme 5 Energy

Deliverables 3.3 and 3.5

Lifecycle assessment and environmental assessment

Name of the coordinating person:

Mr. Raffaello Garofalo Coordinator email: ebb@ebb‐eu.org Coordinator phone: +32 2 7632477 Coordinator fax: +32 2 7630457

Disclaimer: the views expressed in this document are purely the authors' own and do not reflect the views of

the European Commission

REV Date Organisation Beneficiaries involved

Dissemination level

Rev 0 29/03/2011 Dongxu Xu, Raphael Slade, Ausilio Bauen

IMPERIAL BGU, EBB, IMPERIAL, IMIC, NE

PU

Rev 1 10/05/2011 F. Gabriel Acien UAL UAL PU

Rev 2 15/05/2011 Dongxu Xu, Raphael Slade, Ausilio Bauen

IMPERIAL IMPERIAL PU

Rev3 10/06/2011 Benoit Queguineur, Jessica Ratcliff

ISC ISC PU

FINAL 15/06/2011 Dongxu Xu, Raphael Slade, Ausilio Bauen

IMPERIAL IMPERIAL PU

AQUAFUEL FP7 – 241301‐2

Coordination Action

FP7‐ENERGY‐2009‐1

1

Table of contents

1. Introduction ....................................................................................................... 9

1.1 Objective and structure of this Report................................................................... 9

1.2 The Aquafuels project .......................................................................................... 10

2. An introduction to algae cultivation and use .................................................... 11

2.1 Algae Strains ......................................................................................................... 12

2.1.2 The US Aquatic Species Program (ASP) ....................................................... 14

2.2 Micro‐algae Production Systems: Raceway Ponds and Photo‐bioreactors.......... 14

2.2.2 Open Pond Systems..................................................................................... 15

2.2.3 Closed Systems ............................................................................................ 16

2.3 Recovery of Biomass: harvesting.......................................................................... 21

2.3.2 Lipid and product extraction ....................................................................... 22

2.4 Conversion to Biofuels.......................................................................................... 23

2.5 Biomass productivity ............................................................................................ 24

2.5.2 Solar Conversion Efficiency.......................................................................... 25

2.5.3 Lipid biosynthesis and oil producing algae strain selection ........................ 27

3. Life Cycle Assessments of algae biomass production ........................................ 28

3.1 Introduction to Life Cycle Assessment ................................................................. 28

3.2 System Boundaries ............................................................................................... 30

3.2.2 LCA boundary selection: EU Guidelines ...................................................... 30

3.2.3 LCA boundary selection: RMEE method ..................................................... 30

3.2.4 Allocation guidelines ................................................................................... 31

3.2.5 Impact Assessment...................................................................................... 31


Coordination Action


2

4. Review of micro‐algae LCA ............................................................................... 33

4.1 Functional Unit ..................................................................................................... 34

4.2 System boundaries ............................................................................................... 35

4.3 Allocation strategies ............................................................................................. 35

4.4 Sources of data ..................................................................................................... 36

4.5 Algae composition and strain assumptions.......................................................... 37

4.6 Productivity assumptions ..................................................................................... 38

4.7 Global Warming Potential .................................................................................... 39

4.8 Other critiques levied at algae LCA....................................................................... 40

4.9 Conclusions on the existing LCA studies............................................................... 41

5. Meta‐analysis of micro‐algae production systems ............................................ 42

5.1 Meta‐model approach and assumptions. ............................................................ 42

5.1.2 Meta‐model system description and boundaries ....................................... 43

5.1.3 Functional Unit and basis for comparison................................................... 45

5.2 Results .................................................................................................................. 46

5.3 Conclusions........................................................................................................... 54

6. Environmental impacts of micro‐algae production ............................................56

6.1 Water Resources................................................................................................... 56

6.2 Land Use ............................................................................................................... 57

6.3 Nutrient and Fertilizer Use ................................................................................... 58

6.4 Carbon fertilisation............................................................................................... 58

6.5 Fossil Fuel Inputs .................................................................................................. 59

6.6 Eutrophication ...................................................................................................... 59

6.7 Genetic Modified Algae ........................................................................................ 60

6.8 Algal toxicity ......................................................................................................... 61


Coordination Action


3

6.9 Conclusions........................................................................................................... 61

7. Review of macro‐algae LCA ...............................................................................62

8. Environmental impacts of macro‐algae production .......................................... 66

8.1 Land use and near‐shore area use........................................................................ 66

8.1.2 Cultivation at sea......................................................................................... 66

8.1.3 Tank based cultivation on land.................................................................... 67

8.1.4 On‐shore facilities for sea and tank cultivation, wild harvest and bloom

harvest……….................................................................................................................... 67

8.2 Use of Near‐shore/Off‐shore space...................................................................... 67

8.3 Freshwater Use..................................................................................................... 68

8.4 Fertiliser and nutrients ......................................................................................... 68

8.4.2 Cultivation at sea:........................................................................................ 68

8.4.3 Land‐based tank cultivation ........................................................................ 70

8.5 Macro‐algal Domestication and Genetic Engineering.......................................... 71

8.6 Ecosystem Effects ................................................................................................. 72

8.6.2 Cultivation at Sea......................................................................................... 72

8.6.3 Land‐based Tank Cultivation ....................................................................... 73

8.6.4 Wild Harvest................................................................................................ 73

8.6.5 Harvest of blooms ....................................................................................... 74

8.7 Environmental Contamination ............................................................................. 74

8.8 Conclusions........................................................................................................... 75

9. Conclusions and recommendations .................................................................. 75

9.1 Conclusions for micro‐algae LCA, and environmental impacts ............................ 75

9.2 Conclusions for macro‐algae LCA and environmental impacts ............................ 77


Coordination Action


4

10. References ........................................................................................................78

11. Annex 1: Heterotrophic Microalgae...................................................................87

12. Annex 2: Review of existing micro‐algae LCA Studies........................................ 89

12.1 Kadam 2001/2 ..................................................................................................... 89

12.1.2 Functional Unit and System Boundaries ..................................................... 89

12.1.3 Source of Data ............................................................................................. 89

12.1.4 Process......................................................................................................... 90

12.1.5 Results ......................................................................................................... 91

12.1.6 Discussion.................................................................................................... 91

12.2 Lardon et al. 2009............................................................................................... 91


12.2.3 Source of Data ............................................................................................. 92

12.2.4 Process......................................................................................................... 92

12.2.5 Results ......................................................................................................... 93

12.2.6 Discussion.................................................................................................... 95

12.3 Clarens et.al. (2010) ........................................................................................... 95


12.3.3 Source of Data ............................................................................................. 96

12.3.4 Process......................................................................................................... 96

12.3.5 Results ......................................................................................................... 97

12.3.6 Discussion.................................................................................................... 98

12.4 Jorquera et.al. (2010) ....................................................................................... 100

12.4.2 Functional Unit and System Boundaries ................................................... 100

12.4.3 Results ....................................................................................................... 101

12.4.4 Discussion.................................................................................................. 101


Coordination Action


5

12.5 Sander& Murthy (2010) ................................................................................... 102


12.5.3 Source of Data ........................................................................................... 102

12.5.4 Process....................................................................................................... 103

12.5.5 Results ....................................................................................................... 103

12.5.6 Discussion.................................................................................................. 106

12.6 Stephenson et.al. (2010) .................................................................................. 106


12.6.3 Source of Data ........................................................................................... 107

12.6.4 Process....................................................................................................... 107

12.6.5 Results ....................................................................................................... 108

12.6.6 Discussion.................................................................................................. 110

12.7 Campbell et.al. (2010) ...................................................................................... 110


12.7.3 Source of Data ........................................................................................... 110

12.7.4 Process....................................................................................................... 111

12.7.5 Results ....................................................................................................... 111

12.7.6 Discussion.................................................................................................. 113

13. Annex 3: Expert Stakeholders participating in this study .................................114

14. Annex 4: Questionnaire...................................................................................115

15. Annex 5: Assumptions of Normalized Modelling .............................................118

16. Annex 6: Example of data normalization .........................................................119


Coordination Action


6

List of Figures

Figure 2.1 : A raceway pond and farm .................................................................................... 16

Figure 2.2: Tubular photobioreactor system & Flat plate Photobioreactor............................ 19

Figure 2.3: Algal biomass conversion strategies...................................................................... 23

Figure 2.4: Yearly sum of global solar irradiance averages over the period of 1981 to 2000. 26

Figure 2.5: World map of algae biomass productivity ............................................................ 26

Figure 3.1: The analytical stages in Life Cycle Assessment ..................................................... 29

Figure 5.1: Algae LCA meta‐modeling approach..................................................................... 43

Figure 5.2: Description of meta‐model process ...................................................................... 44

Figure 5.3: Definition of Net Energy Ratio (NER) .................................................................... 45

Figure 5.4: NER Biomass production: comparison of published values with normalised values

for algal biomass production. ............................................................................... 47

Figure 5.5: Net Energy Ratio for biomass production in raceway ponds: comparison of

published values with normalised values............................................................. 49

Figure 5.6: Net Energy Ratio for biomass production in photobioreactors PBRs: comparison

of published values with normalised values for algal biomass production.......... 50

Figure 5.7: Illustrative estimates for carbon dioxide emissions from algal biomass production

in raceway ponds.................................................................................................. 51

Figure 5.8: Illustrative estimates for carbon dioxide emissions from algal biomass production

in photobioreactors PBRs. .................................................................................... 52

Figure 5.7: Net Energy Ratio for biomass and lipid production in raceway ponds: comparison

of normalised values............................................................................................. 53

Figure 5.8: Net Energy Ratio for biomass and lipid production in PBRs: comparison of

normalised values................................................................................................. 54

Figure 10.1: General system boundaries for the comparison of electricity production via coal

firing vs. coal/algae co‐firing in Kadam’s study .................................................... 89


Coordination Action


7

Figure 10.2: Simplified flow diagram for microalgae production in Kadam’s study ............... 90

Figure 10.3: Process chain overview from Lardon’s study ...................................................... 93

Figure 10.4: Comparison of impacts categories in Lardon et al ’s study................................. 95

Figure 10.5: Schematic of system considered in Clarens study .............................................. 97

Figure 10.6: System Process in Jorquera’s study................................................................... 100

Figure 10.7: Process flow diagram from Sander & Murthy’s study ...................................... 104

Figure 10.8: Energy and Emissions associated with unit process

in Sander & Murthy’s study................................................................................ 105

Figure 10.9: Process chain for production of 1 ton biodiesel in Stephenson et al’s study ... 108

Figure 10.10: LCA results for base case production of biodiesel

from Stephenson et al’s study ............................................................................ 109


Coordination Action


8

List of Tables

Table 2.1 : Overview on commercially produced micro‐algae ................................................ 13

Table 2.2: Overall Comparison of Open versus Closed Systems.............................................. 18

Table 2.3: Illustrative energy requirements of a tubular photo‐bioreactor’s design .............. 19

Table 2.4: Advantages and disadvantages of alternative closed photobioreactor designs..... 20

Table 4.1: LCA studies on algae derived fuels ......................................................................... 33

Table 4.2: Function units used in the LCA studies ................................................................... 35

Table 4.3: Algae composition assumption in LCA studies ....................................................... 38

Table 4.4: Algae productivity assumptions used in LCA studies.............................................. 38

Table 4.5: Overview of Global Warming Potential claims in algae biomass LCA..................... 40

Table 10.1: Most important material and energy flows generated by the production

of 1kg of biodiesel from Lardon et al’s study ....................................................... 94

Table 10.2: Life cycle burdens of Algae, Corn, Canola, and Switch grass in Virginia............... 98

Table 10.3: Comparative analysis of three cultivation methods from Jorquera’s study ....... 101

Table 10.5: GHG emissions (kg CO2 equivalent) for 1 ton km truck use ............................... 111

Error! No table of contents entries found.


Coordination Action


9

1. Introduction

1.1 Objective and structure of this Report

This report explores the environmental impacts of algae production and the merits of Life

Cycle Assessment (LCA) as a tool for examining the future environmental performance of

transport fuels produced from algal biomass. Specifically, this report examines:

• The current production methods and future possibilities of using algae to produce

biofuels.

• The current status of Life Cycle Assessments of algae derived fuels that are available in

the academic literature; the strengths and weakness of these studies are assessed in

detail.

• The environmental impacts from micro‐algae and macro‐algae cultivation

The report is structured as follows:

• Chapter 1 describes the objectives and structure of the report, and introduces the

AquaFUELS project of which this research is part.

• Chapter 2 introduces the basic concepts of algae cultivation and processing and

reviews the options for cultivation, harvesting and biofuel production.

• Chapter 3 describes the principles of life cycle assessment and the alternative

approaches to setting system boundaries, and allocating impacts to products.

• Chapter 4 presents an analysis of the micro‐algae LCA that are available in the

academic literature. Strengths and weaknesses are identified and the studies are

critiqued. This critique draws on both literature sources and data gathered from expert

stakeholders.

• Chapter 5 presents a meta‐model of the energetics of micro‐algae production. The

data presented in the LCA studies is re‐analysed and normalised to permit a

comparison of the alternative production systems in terms of 1) the energy produced,

and 2) the energy required to construct and operate the system.

• Chapter 6 reviews the major environment impacts which could influence sitting

decisions for micro‐algae cultivation.


Coordination Action


10

• Chapter 7 reviews LCA for macro‐algae

• Chapter 8 review the major environmental impacts associated with macro‐algae

cultivation.

• Chapter 9 presents overall conclusions and recommendations.

1.2 The Aquafuels project

The work presented in this report was undertaken within the context of an EU sponsored

project: Aquafuels (AquaFUELs, 2010). This project aims to bring together and co‐ordinate

existing knowledge, and to establish the state of the art for research, technological

development and demonstration activities regarding the exploitation of algal biomass for 2nd

generation biofuels production. A secondary objective of the project is to put robust and

credible information about algae into the public domain, and thereby counter some of the

more extravagant claims that have been expressed in the media.


Coordination Action


11

2. An introduction to algae cultivation and use

Algae are a large and diverse group of plant‐like aquatic organisms which range from multi‐

cellular macroalgae – e.g. seaweeds such as giant kelp, which can grow up to 60m – to

unicellular microalgae – as small as 3µm. Depending on the species, algae can be farmed

(algaculture) in either freshwater or saline conditions (Carlsson, et al., 2007, Schenk, et al.,

2008). Most algae are photoautotrophic, converting solar energy into chemical forms through

photosynthesis and a variety of biochemical pathways. However many species display some

degree of heterotrophism, utilizing organic carbon as their primary source of carbon and

energy (Barsani and Gualtieri, 2006). Heterotrophic algae are reviewed in Annex 1. The

mechanisms of algal photosynthesis are very similar to photosynthesis in higher plants and

their products are molecularly equivalent to conventional agricultural crops (Graham and

Wilcox, 2000). Although both micro and macro‐algae are of interest within the context of the

Aquafuels project, there is an emphasis within this report upon micro‐algae. The reason for

this is that despite macro‐algae having being cultivated on a larger scale globally than micro‐

algae, there are few assessments of the impacts in the scientific literature.

Microalgae are a broad group of unicellular and simple multi‐cellular photosynthetic

microorganisms that lack complex cell structure and organization. As a source of biomass for

biofuels, potential advantages of algae include:

• high photosynthetic yields (up to a maximum of 5‐6% conversion of light c.f. 1‐2% for

the majority of terrestrial plants);

• the ability to grow in fresh, salt and waste water;

• high overall oil content (dependent on species and growth conditions);

• ability to produce non‐toxic and biodegradable biofuels as well as high concentrations

of commercially valuable compounds such as proteins, carbohydrates, lipids and

pigments;

• the ability to be used in conjunction with wastewater treatment;

• possibility of cultivation on unproductive desert land, thereby reducing competition

for agricultural land.

Regarding biofuel production, microalgae can provide different types of biofuels, including:

methane (produced by anaerobic digestion of algal biomass); biodiesel (from algal fatty acids);


Coordination Action


12

ethanol (produced by fermentation of starch); and hydrogen (produced biologically) (Chisti,

2007, Wang, et al., 2008).

2.1 Algae Strains

The final products from algae aquaculture are determined by the species, strain, and growth

conditions. The US Aquatic Species Program (ASP) looked at over 3000 species; including

strains that can exist in myriad of different environments. One point they focused on was

getting, and then growing, a strain in its native environment, as this was considered to have a

greater chance of success (Sheehan, 1998b). The factors which make an algal strain more

suitable for biofuel production include the following properties:

• a high lipid productivity;

• a high photosynthetic efficiency;

• robustness to growth environment (and in particular the ability to survive the shear

stress from mixing);

• be able to withstand or dominate wild strains in the event of contamination;

• have a high CO2 sinking capacity;

• able to grow in a variety of temperatures and seasons;

• be able to provide (valuable) co‐products;

• be able to self flocculate, or display some of those characteristics (Brennan and

Owende, 2010).

There are currently no algal strains that meet all these criteria. There may also be

additional requirements depending on the location.

The dominant species currently in commercial production are Isochrysis, Chaetoceros,

Chlorella, Arthrospira (Spirulina) and Dunaliella (Carlsson, et al., 2007). An overview of

commercially grown algae is shown in Table 2.1.


Coordination Action


13

Table 2.1 Overview on commercially produced micro‐algae

Source: (Pulz and Gross, 2004)


Coordination Action


14

2.1.2 The US Aquatic Species Program (ASP)

The United States National Renewable Energy Lab (NREL) initiated its Aquatic Species

Program (ASP) in 1978. This programme undertook comprehensive research on algae derived

fuels. After 20 years of research, and 3000 algae strains screened, the project terminated at

1996 due to low oil prices in that decade (Sheehan, 1998b). The overall conclusion for the ASP

project was that low cost production of biofuels from algae was not likely to be feasible within

short or medium term. Although the final report from NREL indicated that the biodiesel from

algae would only become cost effective if conventional diesel prices rose to twice the 1998

levels which was 27 U.S. dollar per barrel.

Factors that have contributed to the renaissance of interest in algae derived fuels, include

policies at regional and national levels, concerns about the security of supply of fossil fuels,

and sustained high oil prices.

2.2 Micro‐algae Production Systems: Raceway Ponds and Photo‐bioreactors

Algal production systems may be classified as either photoautotrophic or heterotrophic.

Photoautotrophic systems use light as the energy source1, while heterotrophic production use

organic substances (such as glucose) to provide the energy the algae require. Some algae

strains can combine these in a mixotrophic process. Currently, the dominant method is

photoautotrophic production of algae as it is the most economically and technically viable,

and shall be the only method discussed here (Brennan and Owende, 2010).

There are two alternative methods for growing photoautotrophic algae: Open Systems

(Raceway Pond System) and Closed Systems (Photo‐bioreactors (PBRs). These systems, and

their variations, are discussed below. An overall comparison is presented in Table 2.4.

1 The reaction in photosynthesis can be summarized as 6CO2 + 12H2O + photons C6H12O6 + 6O2 + 6H2O. Eight photons

must be absorbed to fix one CO2 and two H2O molecules, yielding one base carbohydrate (CH2O) molecule. As the average

energy of “photosynthetically available radiation (PAR) photons is around 217 kJ (accounting for about 43% of incident

sunlight on the Earth’s surface), while the energy content of a single carbohydrate (CH2O) is about 467 kJ/mol, it follows that

the maximum efficiency is roughly 11.6%. In actuality, most plants only have 0.5‐2% efficiency, due to other limitations such

water and nutrient availability, as well as an excess or lack of sunlight, while algae can have as much as 3‐8% efficiency

(Lardon et al., 2009; Vasudevan & Briggs, 2008).


Coordination Action


15

2.2.2 Open Pond Systems

Currently, the majority of commercially grown algae are produced using open pond

systems. These can include natural water bodies such as lakes, lagoons, and ponds, or artificial

systems such as raceway ponds. The latter is the most common, and has been used since the

1950’s. In a typical raceway pond the area is divided into a rectangular grid containing a closed

loop oval shaped channel ~0.2m deep (shallower systems are not commercial, but have been

reported for research purposes. Ponds have to be kept shallow in order to allow the sunlight

to penetrate the water. In most designs some form of mixing and circulation equipment is also

required to stabilize algae growth and prevent sedimentation. Examples of a typical Raceway

pond and an Open Pond farm are shown in Figure 2.1. One of the largest of these types of

systems is the Werribee wastewater treatment plant in Melbourne, Australia, which is 11,000

ha and relies on gravitational flow (Brennan and Owende, 2010, Schenk, et al., 2008).

Raceway ponds that need extra infrastructure (e.g. a paddle wheel) for mixing and tend to

be more expensive to construct than a simple gravity led ponds, as the construction material

(i.e. concrete or compacted earth) has to be able to withstand the shear stress from mixing.

For an Open Pond Raceway, a typical harvest‐growth‐harvest cycle is around 4 days (Sander

and Murthy, 2010).

A comprehensive study undertaken by the US Department of Energy with a variety of

different strains showed that algae selected on the basis of laboratory results would not

compete as well as those that spontaneously colonized and subsequently dominated the

pond, although these adventitious species may not have all the desirable biofuel properties

(Sheehan, 1998b). One method to overcome this is to carefully cultivate and select species

that can dominate the system, such as using the local species, or extremophiles that can

survive well in very particular environments (e.g. extreme temperatures, pH, or salinity). For

example, Spirulina thrives at high pH levels (9 – 11.5), while Dunaliella salina grows well in

saline water (Schenk, et al., 2008). Another option is to cultivate the algae in greenhouse

conditions to control the temperatures and limit contact with contaminating species (Hase,

2000).

Their main advantages of raceway pond systems are that they are relatively easy to operate

and maintain, chiefly due to the simplicity of the design, and have a low energy requirement.

The main is disadvantage is that since they are open to the air, there is higher evaporative

losses and lower utilization of the available CO2 due to diffusion to the atmosphere, which can

lead to changes in the composition of the growth medium harmful for the algae. A comparison


Coordination Action


16

of open versus closed systems is provided in Table 2.2 (Schenk, et al., 2008, Brennan and

Owende, 2010, Chisti, 2007, Oilgae, 2010).

Figure 2.1 A raceway pond and farm

Source: (Sheehan, 1998b)

2.2.3 Closed Systems

The other option for cultivating algae is using a closed system, or photo‐bioreactor (PBR).

These systems have gained in popularity as more high‐value products have been produced

from algae. PBRs tend to be more complex and expensive than open systems, but allow for

better control of the algae culture environment: they can prevent contamination and can be

successfully used to cultivate single species; operate at high biomass concentration; can be

erected over any open space; offering better control of the temperature; and reduce water or

CO2 loss (Amin, 2009, Chisti, 2007, Pulz, 2001a).

Tubular photobioreactor consists of an array of transparent tubes (either plastic or glass) of

0.1m or less in diameter (to allow the sunlight to penetrate the dense medium and allow for

high biomass productivity). In a typical system the micro‐algal broth is circulated round the

system from a central reservoir, which also serves to degas the medium (preventing oxygen

accumulation), harvest the broth, and introduce new broth. Sedimentation is prevented by

either mechanical or airlift pumps. The latter is more inflexible, but does allow CO2 and

oxygen to be exchanged between the medium and the gas. The system is continuously mixed

to prevent sedimentation, even at night when no growth occurs (Amin, 2009, Brennan and

Owende, 2010, Chisti, 2007).


Coordination Action


17

In order to maximize production, incident light needs to be diluted over the surface of the

reactors. This prevents a small area of algae from being oversaturated with light. One of the

simplest approaches to sunlight dilution is to orient photo‐bioreactors vertically, instead of

horizontally, to catch the sunlight over a large surface area (Benemann, 2010a). More

generally, the tubes can be laid out horizontally, vertically (as shown in Figure 2.2), laid out

north to south to maximize solar collection, or coiled around a central support structure. The

ground may also be covered by white plastic to increase the reflectance.

Minimizing the auxiliary energy demand is also an important design parameter. In most

PBR designs energy is required for pumping and mixing to ensure good mass transfer of CO2

andO2. Mixing also helps to prevent the cells from staying too long in dark or bright zones of

the reactor, something that can reduce productivity. Energy may also be needed to cool the

reactors. Pumping energy requirements can be reduced by minimizing the hydrodynamic

pressure of the system. This can be achieved by increasing the diameter of the tubes

(something that may also reduce the overall cost), but the diameter cannot be increased too

far; otherwise light may not be able to penetrate the core of the tube. From a design

perspective there is also a limit on the length of a continuous tube, as the pH may vary within

the system, CO2 may be depleted, and most importantly, photosynthesis produces oxygen

which can inhibit algal growth (Brennan and Owende, 2010, Chisti, 2007, Patil, et al., 2008).

Tubular systems require periodic cleaning, and this increases the operational cost and

water demand. They are also more expensive than raceway ponds due to the higher

infrastructure costs. Illustrative energy requirements of a tubular PBR are shown in Table 2.3.

The comparative advantages and disadvantages of alternative closed photobioreactor designs

are outlined in Table 2.4.


Coordination Action


18

Table 2.2: Overall Comparison of Open versus Closed Systems

Source: (Pulz, 2001b)


Coordination Action


19

Table 2.3: Illustrative energy requirements of a tubular photo‐bioreactor’s design

Total incident solar energy I=150w/m2

Photo conversion efficiency (PCE) PCE=5%

Auxiliary energy demand 50W/m3

Areal water coverage 50L/m2

Areal auxiliary energy demand 2.5W/m2

Source: (Lehr,2009)

Figure 2.2: Tubular photobioreactor system & Flat plate Photobioreactor

Source: (Chisti, 2008a)

There are many alternative Tubular Photo‐bioreactor designs, including flat plate, annular

or column PBRs. Flat plate PBRs increase the surface area of illumination and allow for high

density of cells over a thin layer. Column Photobioreactors offer better control and volumetric

mass transfer rates, and are aerated from the bottom. Their performance is equal to or better

than tubular Photobioreactors. Flat plate systems were researched a lot in the early days,

while column reactors are receiving a lot of attention now, although both systems are still in

pilot scale(Brennan and Owende, 2010).


Coordination Action


20

Table 2.4: Advantages and disadvantages of alternative closed photobioreactor designs

PBR System Advantages Disadvantages

Tubular photobioreactor

• Large illumination surface area• Suitable for outdoor cultures • Relatively cheap • Good biomass productivities • Allows culture of single species• More control and accurate

addition of nutrients and water

• Higher energy requirements • Some degree of wall growth • Fouling • Requires large land space • Gradients of pH, dissolved

oxygen and CO2 along the tubes

• Demonstrated at pilot scale but not scaled up, not commercial

Flat plate photobioreactor

• High biomass productivities • Easy to sterilize • Low oxygen build‐up • Readily tempered • Good light path • Large illumination surface area• Suitable for outdoor cultures

• Difficult to scale‐up • Difficult temperature control • Small degree of hydrodynamic

stress • Some degree of wall growth • Only at pilot scale

Column photobioreactor

• Compact • High mass transfer • Low energy consumption • Good mixing with low shear

stress • Easy to sterilize • Reduced photo‐inhibition and

photo‐oxidation

• Small illumination area • Expensive compared to open

ponds • Shear stress • Sophisticated construction • Only at pilot scale

Source: (Schenk, et al., 2008, Brennan and Owende, 2010, Chisti, 2007, Oilgae, 2010)


Coordination Action


21

2.3 Recovery of Biomass: harvesting

The choice of harvesting method will depend on the type of species involved (i.e. size, density,

etc.), the quantity that needs to be processed and the desired final product. Generally, there

are two main steps:

• Bulk harvesting – separate the biomass from the broth to achieve a slurry with 2‐7% solid

content (involving a concentration factor of 100‐800) (e.g. flocculation, gravity

sedimentation).

• Thickening – concentrate the slurry (i.e. centrifugation, filtration, ultrasonic aggregation). This is generally the more energy intensive step (Brennan and Owende, 2010, Molina

Grima, et al., 2003).

The harvesting process can be highly energy intensive and be quite complex due to the

relatively small size of some microalgal cells (3‐30µm diameter), as well as the relatively dilute

nature of the algal broth (it can be less than 0.5 kg / m3). Some species are easier to harvest

than others, for example, Spirulina (which is 20‐100µm long) can be harvested relatively easily.

Overall, harvesting can contribute as much as 20‐30% to the final cost of production. Notably,

the cost of recovery from photobioreactors may be significantly smaller than from raceways

because the biomass concentration is greater (Brennan and Owende, 2010, Chisti, 2007,

Molina Grima, et al., 2003).

Flocculation is an effective method to aggregate the cells and increase the effective

‘particle’ size, which facilitates the downstream processing. It is important to choose

flocculants that are non‐toxic, effective in low concentrations and will not increase the

amount of downstream processing. Micro‐algal cells generally have a negative charge so

multivalent metal salts are often effective coagulants (i.e. Ferric Chloride (FeCl3), Aluminium

Sulphate (Al2(SO4)3, alum) and Ferric Sulphate (Fe2(SO4)3)). Alum is already widely used in

wastewater treatment. Other options include Polyferric Sulphate (PES), pre‐polymerized metal

salts, or cationic polymers (polyelectrolyte’s) (Molina Grima, et al., 2003)

Filtration is a relatively slow process, but may be a feasible option for low value products

where a higher level of moisture is acceptable. Conventional filtration can be used for larger

algal species, while membrane or ultra‐filtration may be necessary for smaller species. For low

volumes of broth, filtration may be the more economically sound option.

Gravity sedimentation can be used for larger species, but centrifugation is usually the

preferred method of recovery. It is a more energy intensive method, but is also faster and can


Coordination Action


22

handle larger volumes. It also requires more maintenance and has higher costs, but can

increase the slurry concentration by up to 150 times and be up to 95% efficient (Brennan and

Owende, 2010, Molina Grima, et al., 2003, Oilgae, 2010).

2.3.2 Lipid and product extraction

Once harvested, the biomass has to be processed rapidly as it is easily perishable. At this

stage it is typically 5‐15% solid content, but can perish in only a few hours. Algal biomass tends

to have a water content of 80‐90% and low energy density, which, along with the inferior heat

content, makes the biomass harder to use for heat and power generation, necessitating pre‐

treatment (Patil, et al., 2008). This can be achieved by dehydration or drying, which are the

most common methods to treat the slurry, although more expensive than just mechanical

dewatering. Drying methods include: spray drying, drum drying, freeze‐drying, or solar drying.

The choice depends on the desired product. Solar drying for example is the cheapest and

simplest option, but also takes the longest, while freeze and spray drying are expensive and

can cause damage to the cells (although freeze‐drying may help extraction of oils). It is

important to establish a balance between the drying efficiency and cost effectiveness, as well

as the impact on the final product. Temperatures greater than 60°C, for example, can decrease

the lipid yield or denature other components in the biomass (Brennan and Owende, 2010,

Molina Grima, et al., 2003).

To extract the contents of the cells, cell disruption is often necessary. This can be done

mechanically or chemically. Mechanical methods can include using high pressure

homogenisers, bead mills (agitation with glass or ceramic beads), or ultra‐sonication (for small

scale only). Chemical methods include using organic solvents (e.g. hexane), or supercritical

fluid extraction, and may prevent the necessity of drying the cells first (saving a lot of the

energy and costs), but may present health and safety issues (Molina Grima, et al., 2003,

Oilgae, 2010). It is currently commonly done using the same approximate method as for

soybean extraction (hexane solvents), so the algal biomass should have approximately 9 – 11%

moisture content, but this is still the subject of much discussion (Sheehan, et al., 1998a). The

choice of extraction method depends on the final product.


Coordination Action


23

2.4 Conversion to Biofuels

There are multiple strategies to convert algae to energy, depending on the choice of final

product (i.e. electricity, biofuel, etc.) Some of the strategies that can be employed are

illustrated in 2.3.

Figure 2.3: Algal biomass conversion strategies

Source: (Brennan and Owende, 2010)

Many of the methods of processing the algal oil/biomass are developed from those

conventionally used for other systems (i.e. the conversion of vegetable oil to biodiesel or

soybean conversion). Most of these techniques are well established and researched, such as

transesterification – a chemical reaction between triglycerides and alcohol (such as methanol

or ethanol) in the presence of a catalyst (such as sodium hydroxide) to produce biodiesel and

the by‐product glycerol. Glycerol also has commercial value. The final output is similar to

diesel, and the process is relatively simple, so this has been a favoured method for conversion

of vegetable oils (Demirbas, 2009).


Coordination Action


24

Some methods do not require the oil to be extracted, and can convert the whole biomass

into fuels (such as pyrolysis, gasification, and anaerobic digestion). Other methods, such as the

Fischer‐Tropsch method for biomass gasification could also provide the heat needed for the

drying phase (since it is an exothermic reaction). Newer and lower cost methods are emerging

as well, such as the extraction and simultaneous transesterification of oils using supercritical

ethanol or methanol. Some work at high temperatures (such as pyrolysis or direct

combustion), while others (such as anaerobic digestion) work at room temperature. Many

have potential for large scale systems. However, due to the abundance of methods, and the

relative uncertainty of which will be the preferred conversion method, they shall not be

discussed further (Brennan and Owende, 2010, IEA, 2010).

2.5 Biomass productivity

There are several factors to be considered for optimal algae growth. Mostly, they should

recreate as best as possible the natural process of algal growth, which includes:

• Sunlight – this can be the limiting factor as it means no algae is produced at night and

commercial production is limited to areas with a high incidence of solar radiation. This

can be complemented with artificial lighting, but this adds to the energy consumption

of the system, so is used almost exclusively at pilot scale production.

• Nutrients – these include Nitrogen (N), Phosphorus (P), Iron (Fe) and Silicon (Si). Most

algae require N to be in soluble form (such as nitrate, ammonium or urea), although

some can fix atmospheric nitrogen directly. Phosphorus is required in lesser amounts,

but is less readily bioavailable, so must be supplied in excess of the required amounts,

while Silicon is only important to certain species of algae (such as diatoms).

• CO2 – microalgae can fix this from three sources, namely: the atmosphere; discharge

gases from heavy industries (such as power plants); or soluble carbonates. Most

microalgae can utilize considerably higher amounts of CO2 than under normal

conditions (up to 150,000 ppm), so excess carbon can be fed to the system. Typically,

microalgal biomass contains 50% carbon by dry weight, so producing 100 tonnes of

algal biomass fixes 183 tonnes of carbon dioxide.

• Temperature – although there are strains of algae that can survive extreme temperatures,

generally, the optimal temperature is around 20°‐30°C (Brennan and Owende, 2010,

Chisti, 2007, Kadam, 2001).


Coordination Action


25

2.5.2 Solar Conversion Efficiency

Photosynthetic Efficiency (PE) is one of the major factors used to evaluate the growth rate

of terrestrial plants, and is defined as the fraction of light energy which is fixed as chemical

energy during photo‐autotrophic growth. In common with terrestrial plants, there are two

metabolic pathways by which algae fix CO2, known as the C3 (Calvin cycle) or C4 pathways.

Most algae use the C3 pathway which has a maximum theoretical efficiency of ~12% (Tredici,

2010). The maximum that can be practically achieved, however, is ~5%. This is roughly

equivalent to the photosynthetic efficiency of a leaf. The C4 pathway is more efficient (up to

twice the photosynthetic efficiency of C3 plants. (Lundquist, et al., 2010)) and can be found in

diatoms and sugar cane. Many algae strains, however, have evolved to tolerate low light level

and are not only unable to use large amounts of energy at peak light intensities but actually

performed worse under conditions of high exposure to light. This is the principle of light

dilution in photobioreactor design.

Solar radiation is, nevertheless, one of the most important factors influencing algal growth,

and in areas of high insolation (>6 kWh/m2/day), the theoretical maximum production rate for

algae is approximately 100 g/m2/day (Darzins, et al., 2010). The minimum level considered

adequate for algal growth is ~1.5 kWh/m2/day.

The yearly average solar irradiance in different parts of the globe is shown in Figure 2.4.

And, using the 1.5 kWh/m2/day criteria it appears that the majority of the earth‘s land surface

is to be suitable for algae production. To achieve high levels of production throughout the

year, however, it is desirable that there is little seasonal variation. For practical purposes the

suitable locations are those areas where insolation is not less than 3000 hours/yr (average of

250 hours /month) (Necton, 1990, AquaFUELs, 2011b). Most commercial microalgae

production to‐date has occurred in low‐latitude regions. Israel, Hawaii and southern California

are home to several commercial microalgae farms. Figure 2.5 shows the potential yield of

algae biomass at 5% photosynthetic efficiency (Tredici, 2010). Productivity is highest in warm

countries close to the equator where there is little seasonal variation in sunlight levels and

temperatures. (AquaFUELs, 2011b).


Coordination Action


26

Figure 2.4: Yearly sum of global solar irradiance averages over the period of 1981 to

2000.

Source: (Meteotest); database Meteonorm (www.meteonorm.com)

Figure 2.5: World map of algae biomass productivity

Source: (Tredici, 2010); (tonnes ha‐1 year‐1) at 5% photosynthetic efficiency considering an energy content of 20 MJ

kg‐1 dry biomass.

http://www.meteonorm.com/�


Coordination Action


27

2.5.3 Lipid biosynthesis and oil producing algae strain selection

The most common lipids in algal cells are Triacylglycerides (TAG), which are formed from

fatty acids and glycerol. The lipid content in algae can range from 1% to 50% and can vary

greatly with the growth conditions. TAG storage is an important adaptation for photosynthetic

organisms that feast and famine with the diurnal cycle as TAG produced during the day

provides a carbon and energy source for the night.

In eukaryotic algae, lipid content is normally inversely proportional to growth rate; with

lipid is increasing when growth is inhibited by lack of nutrients, such as nitrogen or silicon

(Lundquist, et al., 2010). The fatty acid composition of membrane fluidity and triglyceride

carbon chains can vary in length depending upon the algae species, and environmental

conditions during growth. The growth limitation is not thought to be due to reduced

biosynthetic rates but rather to reduction in other cellular components, leaving a higher

proportion of lipid overall (Lundquist, et al., 2010).


Coordination Action


28

3. Life Cycle Assessments of algae biomass production

This chapter introduces basic concept of Life Cycle Assessment and describes the current

methodologies for conducting Life Cycle Assessment studies.

3.1 Introduction to Life Cycle Assessment

Life‐Cycle Assessment (LCA) is a process formalised by the International Standards

Organisation(ISO, 1997) to evaluate the environmental burdens associated with a products

and processes. LCA seeks to identify and quantify energy and materials consumed and waste

released to the environment, thereby enabling the evaluation and comparison of

environmental improvement options. The assessment includes the entire life cycle of the

product, process, or activity, encompassing extracting and processing raw materials;

manufacturing, transportation and distribution; use, re‐use, maintenance; recycling, and final

disposal” (SETAC, 1993).

In contrast to other environmental management tools, which tend to focus on specific life

stages of a product or process, LCA analyses the entire life cycle, looking up and down the

supply‐chain, from raw material extraction to final disposal. LCA is not site specific and

includes burdens and impacts outside the immediate factory gates. The argument in favour of

the LCA approach is that, whilst traditional environmental assessment tools may overlook the

problem of burden shifting or displacement, LCA ensures that environmental impacts which

have been identified and reduced at one stage of the life cycle are not replaced by other,

possibly greater, environmental impacts elsewhere.

The application of LCA methodology encompasses four phases, Illustrated in Figure 3.1,

below.

• Goal and scope definition: sets the boundaries for the analysis, defines the level of detail

and the functional basis for comparison.

• Inventory Analysis: quantifies emissions, energy and raw materials for each process and

presents these in a process flow chart.

• Impact Assessment: quantifies and groups effects of the resource use and emissions into a

number of environmental impact categories which may be weighted for importance

• Interpretation: reports the results and evaluates the opportunities to reduce the

environmental impact of the product or service.


Coordination Action


29

Figure 3.1: The analytical stages in Life Cycle Assessment

Source: (De Smet, et al., 1996)

Although LCA is a popular tool it has a number of widely recognised limitations:

• The quality of an LCA depends on the quality and availability of accurate data. For

many processes and materials such data does not (yet) exist or is not readily

accessible.

• LCA methods are inherently subjective. Numerous assumptions must be made in

particular relating to the definition of boundaries, the choice of data sources and the

weighting and allocation of impacts

• LCA is a bottom‐up analysis tool which is best used to compare alternative products or

services

• LCA do not take account of rebound effects where environmental and cost efficiency

improvements are cancelled out by greater consumption.

System boundaries, allocation strategies and impact assessment, in particular, merit further

discussion.


Coordination Action


30

3.2 System Boundaries

The system boundaries selected determine what processes and activities are included in

the overall LCA. Many sub‐processes, such as the manufacture of equipment, could potentially

be included and their inclusion/exclusion can strongly influence the outcome of the study.

Various approaches to setting boundaries have been proposed and are described in more

detail below.

3.2.2 LCA boundary selection: EU Guidelines

The EC has published a guide for good LCA practice, and recommends that all the important

processes and activities are included, with only processes of minor importance excluded. It

warns against misleading results due to cut‐off criteria are that are weak, irrelevant, not in

accordance with the intended application, or when they focus on a single flow without due

consideration of all their individual environmental impacts. Another potential problem area is

if there is a lack of proper screening and iteration in the LCA methodology, which may lead to

the exclusion of activities without justifying whether they are significant or not, leading to

misleading conclusions (EC, 2010).

Article 17 in the 2009 Renewable Energy Directive deals with the sustainability criteria for

biofuels, and states that raw materials for biofuels should not come from protected land (e.g.

land with high biodiversity). This implies the system boundaries should, where possible, be

drawn back to include extraction of raw materials from the earth (EC, 2009).

3.2.3 LCA boundary selection: RMEE method

One proposed method to systematically and quantitatively set the system boundaries is the

Relative Mass, Energy, and Economic value (RMEE) protocol. In this protocol, the relevant data

about individual processes is gathered before the system boundary is drawn. A predefined cut‐

off ratio is then applied to the functional unit on the basis of mass, energy and economic

value. Starting with the process units closest to the functional unit, the RMEE ratios are

calculated for each input, and if larger than the cut off ratio, it is included in the system

boundary. This is repeated until all upstream processes are below the cut off threshold. This

helps reduce the subjectivity inherent in LCAs (Sander and Murthy, 2010, Raynolds, et al.,

2000a, Raynolds, et al., 2000b).


Coordination Action


31

3.2.4 Allocation guidelines

Allocation of the various input and output streams credited to any co‐products created is

considered one of the weaknesses of current biofuel LCAs. Co‐ or by‐products are any

products that are obtained from the process in addition to the desired product.

The ISO recommends avoiding allocation if possible, preferring to divide the unit process

into sub‐processes or expanding the system to include the functions of the co‐products. If

allocation is necessary, it should reflect the physical relationship between co‐products (or if

not possible, other relationships such as the economic value), reflecting the way in which the

inputs and outputs are changed, although not necessarily in proportion to simple

measurements such as the mass flow (ISO, 1997, SAIC, 2006).

These guidelines can be broken down into several concrete methods of allocation. One is

the use of direct substitution, where the by‐product (i.e. heat) from a process can be directly

used elsewhere; thereby replacing what would otherwise have been used. This method is

useful when there is a direct use for the by‐product, but in situations when the by‐product is

viewed as waste is less useful. Other methods exist, such as allocation on an arbitrary basis

(i.e. equal value), or on the basis of economic, calorific value or mass. Each has advantages

and disadvantages; for example, allocation based on market value is highly variable over time

and in some cases, where one product far outweighs another, may also be impractical (SAIC,

2006, Stephenson, et al., 2010).

3.2.5 Impact Assessment

Numerous impact categories can be assessed. The most common ones selected are:

• Greenhouse gas emissions:

• Energy use: any energy recovered from waste is historically given as the higher heating

value (HHV) of the materials being burned

• Water use: impact on water, including:

o Water consumed during process


Coordination Action


32

o Waterborne emissions: discharges into any receiving waters after treatment,

measured as a value of the biological oxygen demand (BOD), chemical oxygen

demand (COD), or suspended/dissolved solids

• Solid waste: any waste sent to landfills

• Land use

Economics of the process (Heijungs, et al., 1992):

Generally speaking it is desirable that the outcome should be quantitative given in either a

range of values, or, depending on the information available, cover a variety of different

production methods (Heijungs, et al., 1992, Boguski, et al., 1996).


Coordination Action


33

4. Review of micro‐algae LCA

Despite a high level of interest, no industrial scale processes designed specifically for micro‐

algal biofuel production yet exist. Only a limited number of LCA have been conducted, and,

because of the lack of data from operating plant, they are somewhat speculative. A systematic

review of the literature in late 2010 identified 7 LCA assessments, listed in Table 4.1. For

reference, a detailed description of each study is provided in Annex 2. No life cycle

assessments of macro‐algae production were found.

This chapter reviews the main features of the studies – i.e. choice of functional unit,

boundaries, allocation strategies, etc. A discussion of each of each of these aspects is also

provided. The basis of this discussion is comments and criticisms that have been made in the

literature, and information gathered from stakeholders. Input from stakeholders was elicited

using a questionnaire and semi‐structured interviews. (Stake holders consulted are listed in

Annex 3; the questionnaire used to elicit their input is described in Annex 4.)

Table 4.1: LCA studies on algae derived fuels

Author & year Description

Kadam 2002

This study compares a conventional coal‐fired power station with one in which coal is co‐fired with algae cultivated using recycled flue gas as a source of CO2. The system is based in the southern USA, where there is a high incidence of solar radiation.

Lardon et.al. 2009

This study considers a hypothetical system consisting of an open pond raceway covering 100ha, and cultivating Chlorella vulgaris. Two operating regimes are considered: normal levels of nitrogen fertilisation, and low nitrogen fertilisation. The stated objective was to identify obstacles and limitations requiring further research.

Clarens et.al. 2010

This study compares algae cultivation with corn, switch grass and canola. The study was based in Virginia, Iowa and California in the US, each of which have different levels of solar radiation and water availability five impact categories: energy consumption (MJ), water use (m3), greenhouse gas emissions (kg CO2 equivalent), land use (ha), and eutrophication (kg PO4)


Coordination Action


34

Jorquera et al. 2010

This study compares the energetic balance of oil rich microalgae production. Three systems are described: raceway ponds, tubular horizontal PBR, and flat‐plate PBRs. No specific location was assumed, and the study only considers the cultivation stage and the system energetics

Sander & Murthy 2010

This was a well‐to‐pump study that aimed to determine the overall sustainability of algae biodiesel and identify energy and emission bottlenecks. The primary water source was treated wastewater, and this was assumed to contain all the necessary nutrients except for carbon dioxide. Filtration and centrifugation were compared for harvesting. Lipids were extracted using hexane, and then transesterified.

Stephenson et. al. 2010

This study is a well‐to‐pump analysis, including a sensitivity analysis on various operating parameters. Two systems were considered, a raceway pond and an air‐lift tubular PBR. The location of the study is in the UK, which has lower solar radiation than the other studies.

Campbell et.al. 2010

This study looked at the environmental impacts of growing algae in raceway ponds using seawater. Lipids were extracted using hexane, and then transesterified. The location of this study was in Australia, which has a high solar incidence, but limited fresh water supply.

4.1 Functional Unit

The function units used in the LCA studies are listed in Table 4.2. It can be seen that a diverse

range of units has been used. Although, as (Clarens, et al., 2010) notes, the choice of

functional unit may not be that important for an individual study, as it is only used to assess a

given aspect of a life cycle. Nevertheless, the widespread use of incomparable units prevents

easy comparison between studies. (Benemann, 2010b) recommends a functional unit of “CO2

emissions per gallon of biodiesel or similar biofuel delivered to the plant gate”. Fonseca

(Fonseca, August 2010) goes further and suggests that the calculation, unit capacity and

conditions should be standardized across studies; he recommends the following functional

units: “for assessing the energetic balance ‐ energy produced/1 unit of energy consumed in the

whole process, and for environmental balance – GHG emissions/1 unit of energy produced (of

the whole process, and of each relevant step)”.


Coordination Action


35

Table 4.2: Function units used in the LCA studies

Author & year Functional unit

Note

Kadam 2002

1MW Of electricity

Lardon et.al. 2009

1MJ Of fuel used in a diesel engine (assumed to be identical to other biofuels)

Clarens et.al. 2010

317 GJ Equivalent to the approximate per capita primary energy consumption of one American


100,000 kg Biomass dry weight per annum


1000 MJ Energy in the form of algal biodiesel produced using existing technology at a filling station (equal to 24kg dry algal biomass)


1 Tonne Biodiesel blended with conventional diesel, delivered toa UK filling station and used in an average UK car

Campbell et.al. 2010 0.89 MJ The diesel fuel equivalent to transport one tonne of freight one kilometre in a an articulated truck (the most common form of freight transport in Australia)

4.2 System boundaries

The studies all adopt different approaches to setting the system boundaries. For example,

(Jorquera, et al., 2010) only considers the cultivation phase, whereas Clarens et al. (2010)

considers both cultivation and harvesting. Sander & Murthy (2010) adopt the RMEE protocol

(described in Chapter 4) to define their system boundaries, and this means that they are the

only study to consider the production of algae inoculums. Stephenson et al. (2010) found that

the manufacture of equipment (PVC lining and the PBR tubes) required a large energy input,

yet many of the studies did not include this aspect.

Production processes can only be fairly compared if the system boundaries are the same.

Nevertheless, there may be good reasons for varying the selected boundary. Greenwell

(2010), for example, argues against making a standardized system, as a local consortium in a

developing country will have different considerations (environmental, economic, energetic,

and societal) to a large company in the U.S.

4.3 Allocation strategies

Only two of the studies, Sander & Murthy (2010) and Stephenson et al.(2010), consider how

impacts should be allocated to products and co‐products. Sander & Murthy (2010) show how


Coordination Action


36

the choice of co‐products could have a major impact on the final GHG emissions and

energetics of the system, while Stephenson et al.(2010) demonstrated the potential

importance of using the residual biomass for methane production. Neither deals exclusively

with one form of allocation, although Stephenson et al.(2010) did show that alternative uses

of minor co‐products (glycerol) – which consequently change the allocation method – did not

make a big difference overall. It may be argued that Campbell et al. (2010) should be included

in this list as includes electricity generation from the algal cake. More generally, there are

numerous of previous assessments of biofuels that have shown how important allocation

calculations are on the final outcome (Gnansounou, et al., 2009).

Different experts prefer different methods of allocation and there is little consensus. For

example Clarens (2010b) and Fonseca (2010) prefer market valuation, while Greenwell (2010)

prefers the simplest option.

Benemann (2010b) argues that biofuels should be considered a co‐product of wastewater

treatment and not the other way around, and eventually that biofuels should be a separate

business. Biomass from algae grown for feeds would have a higher market value sold

elsewhere than they would from biofuels, so co‐products would not be generated and there

would be no need to allocate. However, a survey undertaken by the EABA (2010) shows that

many members of the industry are interested in producing energy in conjunction with some

other product.

4.4 Sources of data

There is significant variation in the parameter values used in the studies. The majority are

based on pilot or lab scale values and different methods have been used to estimate how

these values scale to a full size system. Many studies use data which is over 15 years old; for

example, one of the most frequently cited papers is a one by (Benemann and Oswald, 1996),

which discussed the various possibilities for algae production. Other commonly used data

sources for the studies include the EcoInvent database, various LCI databases, and GREET.

One of the major criticisms of the current studies is their lack of transparency about data

sources, and the lack critical thinking about how reliable the sources and assumptions are

(Benemann, 2010b; Greenwell, 2010). For the most part the studies do credit their sources,

however, many refer to their own earlier work (e.g. Campbell et al. bases his system on a

previous paper written in 2009).


Coordination Action


37

Some experts also believe that scope for technical advance is significant. Consequently, the

literature may be outdated and the assumptions unduly negative, or incorrectly chosen. As

Fonseca (2010) says:

“Available options for optimization in each step of the technology are many, but just few

have been analysed in the referred LCA. The negative values some LCA demonstrate for algae

biotechnology do not mirror reality because the initial conditions and technological options

were not correctly chosen”.

Another consideration to keep in mind about the data use is that some of the authors may

be LCA experts, but not experts in the field of algae cultivation. Greenwell (2010) sums this up

by saying:

“[LCA studies] tend to be conducted by either LCA specialists who are not specialists in the

technology, or do not have enough aspects of the process covered. For example, palm oil

would look pretty good by energetic or economic LCA, but societal pressures prevented its

take‐up”.

Clearly, due to the hypothetical nature of the current work, arguments about the values

used are bound to happen, and until an actual system is created, basing the work on pilot

systems with critical thinking of the effects of scaling up is the best that studies can do.

4.5 Algae composition and strain assumptions

The major components of algal biomass are carbohydrates, proteins and lipid. Each species

differs in their physiological composition of these components. The relative proportion of

components will also depend upon the growth regime. The values used in the studies are

shown in Table 4.3. It can be seen that the lipid varies from 17.5‐43%, carbohydrate from 20‐

53%, and protein from 5.5‐32%. These composition assumptions may significantly affect the

final result. It should also be noted that a high lipid content will usually comes at the cost of

lower productivity, and this will also affect the result of the LCA. High lipid content will not

necessarily always be beneficial, however, as Greenwell (2010) argues: “many [LCA] do not

address the problem that biodiesel only can use a very limited range of the oil produced in

algae. [This is usually ignored] because the production chemists usually have no input into the

LCA” (Greenwell, 2010).


Coordination Action


38

Table 4.3: Algae composition assumption in LCA studies

Lardon et al. Kadam

Normal N

Low N

Jorquera et al.

Clarens et al.

Sander & Murthy

Stephenson et al.

Campbelll et al.

Algae strain ND Chlorella vulgaris

Nanno‐ chloropsis

sp. ND ND

Chlorella vulgaris

ND

Lipids 30.0 17.5 30.0 29.6 ‐ 30.0 45.01 43

Carbohydrates 20.0 49.5 52.9 ‐ ‐ 31.0 49.3 ‐

Proteins 32.0 28.2 6.7 ‐ ‐ 37.5 5.5 ‐ 1 Original study assumed 40% TAG, and 5% free fatty acid

ND = Not defined in study

4.6 Productivity assumptions

The productivity estimates used in the LCA studies are compared in Table 4.4. These estimates

illustrate the range of optimism about possible algae growth rates, and describe a range from

26‐112 t/ha/a, with an average of ~54 t/ha/a. For comparison a number of other estimates

that can be found in the literature are also shown (Chisti, 2008b; Griffiths & Harrison, 2009;

Sawayama et al., 1999). These studies indicate a less optimistic maximum annual productivity

of ~60t/ha/a.

It should be noted, however, that some studies report productivity per unit volume rather

than per unit area. Moreover, the length of the growing season assumed is not necessarily

transparent. Consequently, some manipulation of the figures provided in the papers is

required before they can be compared on an equal basis.

From the perspective of conducting comparable LCA in the future, the best choice of unit is

probably the average annual productivity, expressed per unit area.

Table 4.4: Algae productivity assumptions used in LCA studies

Production rate quoted in studyStudy Per day

(g/m2/d) Per annum (t/ha/a)

Normalised annual production ratea (t/ha/a)

2001 17.1 33.17 42.75b Kadam

2002 45.0 104 112.50 Normal N 24.8 ‐ 61.88

Lardon et al Low N 19.3 ‐ 48.13


Coordination Action


39

California 12.9e 47.1 32.26 Virginia 11.0 e 40.2 27.53 Clarens et al.c Iowa 14.1 e 34.5 35.35 Raceway 10.5 e 38.5 26.36 Flat‐plate PBR 27.0 e 98.6 67.50 Jorquera et al. Tubular PBR 25.5 e 92.9 63.64

Sander & Murthy

n/a ‐ ‐ ‐

Raceway 27.4 e 68.49 Stephenson et al. PBR 27.4 e

100 68.49

Low 15.0 54.8 37.50 Campbell et al.

High 30.0 109.6 75.00 LITERATURE

Chlorella vulgaris 16.0 ‐ 40.00 Nannochloropsis 15.0 ‐ 37.50

Griffiths & Harrison

Average 24.0 ‐ 60.00 Chisti 25.00 ‐ 62.50 Sawayama et al.d

B. braunii 4.11 15 10.27

a unless otherwise stated, it is assumed that the quoted annual production rate is for 365 days. The

normalised production rate assumes 250 days of production per year, corresponding to a May to October in a

seasonal country where winter inhibits growth. bNormalised value assumes 100% of culture area is productive, whereas original study assumes only 86% of

area is productive. cClarens et al. varied their production values over the year (the average is given here) dChisti (2008b) questions the validity of this productivity estimate, claiming it is only 16% of current

production in the tropics. e Value calculated from figures presented in the original study.

There is a general consensus among the experts questioned that algae growth rate in terms

of biomass productivity and lipid productivity are far too optimistic and do not take into

account the losses that would occur with scaling up the process. It is assumed that the

productivity values given are based on the year average, for, as Greenwell (2010) points out:

“comparing mean productivity on a given day is not the same as averaging over a whole year”.

4.7 Global Warming Potential

Comparisons made to conventional biofuels and fossil fuels depend on the reference systems

used and direct comparison is not straightforward. An overview of the relative GWP savings

claimed in the reviewed studies is shown in Table 4.5. Caution is required, however, because


Coordination Action


40

the studies use very different boundary and allocation assumptions. Nevertheless, it can

clearly be seen that there is a wide range of estimates ranging from significant savings to

additional emissions.

Table 4.5: Overview of Global Warming Potential claims in algae biomass LCA.

Author & year Reference system CO2 balance Kadam 2002

Electricity from coal firing

‐36.72% (direct injection of the flue gas) ‐2.46% (monoethanolamine (MEA) extraction of CO2 from flue gas

Lardon et.al. 2009

Diesel ‐25%

Clarens et.al. 2010

Corn / canola / switchgrass

+244% / +189% / +233%


NA NA


Gasoline ‐117% (dewatering using filter press)a +14% (dewatering using centrifuge)


Diesel ‐ 78% (Raceway ponds) + 273% (PBR)

Campbell et.al. 2010

Diesel per freight km.Tonne

‐66% ‐122%

a This study assumes that co‐produced algal residue displaces animal feed.

4.8 Other critiques levied at algae LCA

Other, more general critiques levied at algae LCA include concerns about the quality,

representativeness, and application of the studies. As illustrated by the following comments

from expert stakeholders:

• LCA as a very complex tool to use on such a young industry… [Fonseca (2010)] • “LCAs should at least mention the temporary nature of the production pathways in use, or

possibly assess the "representativity gap" of other LCAs performed on experimental pathways but intending to derive findings applicable to algae production for biofuels”. [Vernon 2010.]

• LCA studies cannot so easily be adapted to such a new industry, and do not give a holistic view of the whole sustainability (the “global wrapping”), including the use of land, water, etc. [Leu (2010) ] • From an industry point of view, it is happening the worst possible thing; a pollution of publications on microalgae production LCA which refer to each other and in many cases are careless and get strange conclusions (which are interesting to publish)”. (Vieira, 2010)


Coordination Action


41

• [Referring to the Clarens et.al. 2010 report] The report was based upon obsolete data and grossly outdated business models, and overlooked tremendous improvements in technology and processes across the production cycle. [These] obsolete data and faulty assumptions seriously undermine the credibility of the study’s conclusions. (ABO, 2010)

• “The fundamental problem here is that biofuels are subsidized and therefore we get into the quandary of converting low price fossil fuels into high price subsidized biofuels, which is why we now have to do LCAs. Actually, a better measure may be process sustainability over a 100 year period, in which all non‐renewable inputs are accounted for. For example for algae, use of phosphates has a very small impact on LCAs but a potentially very large impact on sustainability, as phosphate mining is a rapidly depleting resource. If this is not accounted for as a major factor in an LCA, which currently focuses mainly on greenhouse gases, it is not useful”. [Benemann ]

4.9 Conclusions on the existing LCA studies

This review of existing LCA supports the following conclusions:

• The studies reviewed here consider a wide range of conceptual designs, but, with the

exception of the study by (Stephenson, et al., 2010), they all provide only partial

descriptions of algal biofuel production systems. Studies such as (Kadam, 2001) which only

consider the production stage, will naturally provide a more positive energy balance than

studies that include subsequent energy intensive processing steps.

• Comparison is also hindered by the use of inconsistent boundaries and functional units.

• The studies use a range of allocation methods, some of which, it may be argued, are

overcomplicated given the immaturity of the industry.

• It is also evident that, in the absence of commercial production facilities there is little

primary data upon which process assumptions can be based. The production processes

analyzed appear to be assembled from component parts, rather than designed as

integrated systems.

• The validity of some of the results and the usefulness are called into question, by experts

in the field. Prominent causes of concern include the hypothetical nature of the LCA, and

that without more detailed information about the system, the strain used, and the

possible amount of oil that can be extracted, the conclusions that may be drawn from the

existing LCAs are tenuous at best.


Coordination Action


42

5. Meta‐analysis of micro‐algae production systems

Comparing the results of the existing LCA studies is challenging because of the inconsistencies

in boundaries, allocation strategies etc. used in the different studies (Described in Chapter 5).

To enable a comparison of algae production systems in terms of 1) the energy produced, and

2) the energy required to construct and operate the system, an LCA meta‐model was built and

populated using data from the original studies. This chapter describes the meta‐modelling

approach, and the results obtained.

5.1 Meta‐model approach and assumptions.

The objectives of the meta‐model were two‐fold; firstly, to enable a more detailed

examination of the assumptions used in the existing LCA, and secondly, to compare the

studies in terms of the energy produced and consumed. The model was built in Excel using a

simplified, but complete, description of the processes involved in cultivation, harvesting of

algal biomass, and extraction of algal oil.

The modelling approach (shown in Figure 5.1) was undertaken in three stages. Firstly, the

data and assumptions contained in the original studies were identified and transcribed.

Secondly, the units were normalized. Lastly, the process descriptions were normalized to fit a

consistent system boundary, and to allow comparison using a single functional unit. An

overview of the assumptions used to normalize the studies are described below, detailed

assumptions for the normalization of each study are described in Annex 5 & 6.


Coordination Action


43

Figure 5.1: Algae LCA meta‐modeling approach

5.1.2 Meta‐model system description and boundaries

The meta‐model process system is shown in Figure 5.2. In Stage 1, the algae are cultivated

and harvested. At this stage both raceway pond and closed (PBR) systems are included for

comparison. For each study the efficiency with which nutrients and CO2 are captured is based

on the original study. The residence time of the algae strain in the cultivation system – where

algae cell has to accumulate the lipid (TAG) to a certain level for biodiesel production – also

follows the original studies. The boundaries include the manufacture of the principal

equipment (e.g. the PVC lining for the raceway pond systems and system maintenance and

operations).

In Stage 2, the slurry of mature algae cells is transported to a dewatering and drying stage.

The amount of drying required is determined by the oil extraction process (Stage 3). Most of

the LCA studies adopt hexane extraction as they assume algal oil extraction will be very

similar to soybean oil extraction2. This process requires that the paste has to be dried up to a

solid content of ~90% before being processed in oil mill3. In order to achieve this a belt dryer

2 It should be noted that some experts contest this assumption

3 A belt dryer, usually is used for wastewater treatment plant sludge, is assumed as it is one of the less energy

demanding drying process. Heating supplied in the system comes from natural gas combustion.


Coordination Action


44

was chosen as the preferred technology for biomass drying based on data presented in

Lardon et al., (Lardon, et al., 2009).

Stage 3 is oil extraction. The extraction efficiencies are based on the data from original

studies. Co‐products from oil extraction – mainly carbohydrate and protein – are assumed to

be used as feedstock to produce biogas in Stage 4 (Sialve, et al.). This biogas is then used in a

gas boiler to generate heat for the drying process. Any excess gas is converted into electricity,

which is also used for system operations.

Figure 5.2: Description of meta‐model process

1. Harvesting (Raceway ponds / PBR)

2. Dewatering; drying

3. Oil extraction

Algae Lipid

Algae Cake

Algae

Nutrients; Carbon dioxide; Energy

Residue (Carbohydrate,

Protein)

Flue gas; nutrients

Flocculant; Energy

Hexane; Energy

4. Biogas production and

combustion


Coordination Action


45

5.1.3 Functional Unit and basis for comparison

The functional units selected for comparing energetic performance were 1MJ dry algal

biomass and 1MJ algal lipid. Alternative processes were compared in terms of the Net Energy

Ratio (NER) of biomass and lipid production, defined in Figure 5.3. If the NER is greater than

unity, the process consumes more energy than it produces.

Figure 5.3: Definition of Net Energy Ratio (NER)

To calculate the NERBiomass three processes stages are considered: algal biomass cultivation,

drying and dewatering and oil extraction. No co‐product allocation is applied and we assume

that the energy content of the dry biomass is equal to the lower heating value of the algal

biomass specified in the original LCA studies. For those studies that didn’t specify heating

values, estimates were made by summing the heating values of the biomass compositions

given.

Primary Energy Inputs were assumed to include the energy content of the fossil fuel inputs

only; i.e. the embedded energy from the production of the fossil fuel itself is excluded from

the boundary. The energy associated with building the plant was also included (assuming a

20yr lifetime for concrete and 5yrs for PVC (Stephenson, et al., 2010)).

For illustrative purposes, the CO2 emissions for algae biomass cultivation are estimated by

multiplying the primary energy inputs by the default emissions factors described in the EU

renewable energy directive (2009/28/EC)4.

4 The default emissions factors outlined in the renewable energy directive(2009/28/EC) are – diesel: 83.80gCO2.MJ-1; electricity: 91 gCO2.MJ-1; Heat: 77 gCO2.MJ-1. The emissions factor for the embodied energy in fertiliser and for production of PVC lining (in the case of raceway ponds) and tubes (in the case of PBRs) was assumed to be the same as for heat.


Coordination Action


46

The calculation of the NERLipid assumes that co‐products are treated by anaerobic digestion.

Anaerobic digestion is appropriate for processing high moisture content, such as 80%~90%

moisture organic wastes. (Two of the existing LCA studies took anaerobic digestion for algal

residual application)5. It has been estimated that the conversion of algal biomass into

methane could recover as much energy as obtained from the extraction of cell lipid, while

leaving a nutrient rich waste product which could be recycled into a new algal growth

medium.

In the normalized case, we assume anaerobic digestion could convert 60% of the algal

residue to methane (on an energy basis) (Sialve, et al.). This methane is then fed into a gas

boiler to generate heat and offset natural gas demand for the drying process. The efficiency of

the gas boiler is assumed to be 75% (Stephenson, et al., 2010). In the Stephenson et al’s

study, they took homogenization before lipid extraction: this means that the algae biomass

could have higher moisture content in the subsequent processing stage and that not all

residue was required to produce heat for the biomass drying and dewatering process.

Consequently, we applied another scenario to normalize Stephenson et al’s case, assuming

that the gas boiler was replaced by a gas‐engine combined heat and power system (CHP). The

generated heat and electricity is then used to offset fossil fuel used in the production process

(34% efficiency in electricity generation; 41% efficiency in heat generation; overall efficiency is

75%) (Macadam, 2010).

5.2 Results

The seven studies describe eleven alternative processing systems. The primary energy input

for biomass production for each alternative system, before and after normalization, is shown

in Figure 5.4, It can be seen that in all cases the primary energy input for the normalized

process is equal to, or less attractive than, the original case. It is also noticeable that the

closed systems, especially tubular PBR, demonstrate poor energetic performances compared

to raceway ponds.

Six, out of eight, of the raceway pond systems have a Primary Energy Input less than 1,

suggesting that a positive energy balance may be achievable for these systems, although this

benefit is marginal in the normalized case. The normalized Primary Energy Inputs for PBR

systems are all greater than 1. The best performing PBR is the flat‐plate system. This appears

5 Sander and Murthy (2010) and Stephenson et al(2010)


Coordination Action


47

to outperform tubular systems as it benefits from a large illumination surface area and low

oxygen build‐up.

Figure 5.4: NER Biomass production: comparison of published values with

normalised values for algal biomass production.

NC: Normal Cultivation; LN: Low Nitrogen Cultivation; FP: Filter Press; C: Centrifuge; Flat‐plate PBR: Flat‐plate

Photobioreactor; Tubular PBR: Tubular Photobioreactor. RP: Raceway Pond

*=Normalised system boundary

The energy inputs to raceway pond systems are shown in Figure 5.5, expanded to show the

energy input into the different stages.

The three studies where normalisation has the greatest impact are Kadam, Jorquera and

Campbell. Originally these studies only considered the cultivation stage; the addition of drying

and dewatering processes and lipid extraction changes the NER from ~0.05‐0.1 to 0.5‐0.75. For

these studies, even if drying and lipid extraction were excluded, the normalised value for

cultivation is less favourable. This is because the original studies did not include system

construction. (The normalised system also includes transport of fertiliser (data from

Campbell), and the embodied energy in the fertiliser, but these factors are comparatively

insignificant.)

Sander & Murthy use high values for the energy required for algae culture, drying and

harvesting, and these systems will deliver less energy output that they require input. The


Coordination Action


48

original assumptions about the algal species and its productivity are unclear but the data

appears to come from studies completed in the 1980’s, and so may not be representative of

more recent designs.

Stephenson et.al is the only LCA that gives a complete description of the cultivation, and

harvesting process, and so normalisation makes no difference in this case. The energy

demands of the cultivation stage are higher than other studies because the authors assume

more electricity is required at this stage to overcome frictional losses (which they estimate

from first principles). Less energy is required for drying than other studies because, in a

subsequent step, the authors assume the use of an oil extraction process that can accept wet

biomass (homogenisation with heat recovery).

Another source of variation is that each study selects a different composition for the algae

produced and a different productivity for the growth phase; this affects the energy required

per functional unit produced. If the productivity of the algae is assumed to be low, then, all

else being equal, it follows that the energy required to produce 1MJ dry biomass will be

greater (as the mixing requirement per unit time etc. will not be reduced). One complicating

factor is that growing the algae under lower productivity conditions, such as nitrogen

starvation, may allow the algae to accumulate more lipid and so may result in a higher calorific

value for the biomass overall. It is clearly important that productivity and composition values

correspond with one another.


Coordination Action


49

Figure 5.5: Net Energy Ratio for biomass production in raceway ponds: comparison

of published values with normalised values.

0.00

0.50

1.00

1.50

2.00

2.50Ra

ceway Pond

Raceway Pond*

Normal Cultivation

Normal Cultivation*

Low Nitrogen

Low Nitrogen

*

Raceway Pond

Raceway Pond*

Raceway Pond

Raceway Pond*

Raceway Pond

Raceway Pond*

Filte

r Press

Filte

r Press*

Centrifugatio

n

Centrifugatio

n*

Kadam(RP) Lardon(NC) Lardon(LN) Jorquera(RP) Cambell(RP) Stephenson(RP) Sander&Murthy(FP)Sander&Murthy(C )

NER

Biomass

Prim

ary En

ergy Inpu

t(MJ/MJ D

ry Algal Biomass)

Lipid Extraction Biomass Drying and Dewatering Algae Cultivation and Harvesting

NC: Normal Cultivation; LN: Low Nitrogen Cultivation; FP: Filter Press; C: Centrifuge; RP: Raceway Pond


At the cultivation stage, the most important contributions to the energy demand come from

the electricity required to circulate the culture 0.02‐0.79MJ/MJdrybiomass, and the embodied

energy in system construction 0.05‐0.14MJ/MJdrybiomass (8‐70%). The embodied energy

contained in the nitrogen fertiliser ranges from 0.02‐0.09 MJ/MJdrybiomass (6‐40%) (this range

excludes the Kadam study which includes a fertiliser value of 0.05g nitrogen per kg dry algae (a

value that appears infeasably low given that this study assumes the biomass contains >30%

protein).

The energy inputs to PBR systems are shown in Figure 5.6 All the normalised systems

consume more energy than they produce. Biomass drying and de‐watering are

proportionately less important than the energy consumed in cultivation and harvesting. This is


Coordination Action


50

partly because greater algal biomass concentrations can be achieved in PBR systems, and

partly because PBRs consume more energy for cultivation and harvesting than raceway ponds.

In the tubular PBRs, the electricity used to pump the culture medium around the system

and overcome frictional losses accounts for 2.5‐5.0MJ/MJdrybiomass (~86‐92% of the energy

input to the cultivation and harvesting stage) (0.22MJ/MJdrybiomass (22%) for the flat‐plate PBR).

System construction accounts for the majority of the remainder: 0.34‐0.36MJ/MJdrybiomass (6‐

12%).

Figure 5.6: Net Energy Ratio for biomass production in photobioreactors PBRs:

comparison of published values with normalised values for algal biomass

production.

0.00

1.00

2.00

3.00

4.00

5.00

6.00

Flat‐plate PBR Flat‐plate PBR* Tubular PBR Tubular PBR* Tubular PBR Tubular PBR*

Jorquera(FP PBR) Stephenson(T PBR) Jorquera(T PBR)

NER

Biomass

Prim

ary En

ergy Inpu

t (MJ/MJ D

ry Algal Biomass)

Lipid Extraction Biomass Drying and Dewatering Algae Cultivation and Harvesting FP PBR: Flat‐plate Photobioreactor; T PBR: Tubular Photobioreactor

*Normalised system boundary

The carbon dioxide emissions associated with algal biomass production are shown in Figures

5.7 and 5.8. The results essentially mirror the results for the energy consumption and the

majority of emissions are associated with the drying phase and the consumption of electricity.

Notably, emissions associated with algal biomass production in raceway ponds are comparable


Coordination Action


51

with the emissions from the cultivation and production stages of rape methyl ester biodiesel.

Production in PBRs, however, demonstrates emissions greater than conventional fossil diesel.

An important caveat to this analysis is that the carbon emissions are highly dependent on the

emissions factors used for the different energy inputs into the system (and in particular

electricity) and generic factors may not be appropriate in all situations.

Figure 5.7: Illustrative estimates for carbon dioxide emissions from algal biomass

production in raceway ponds

0

50

100

150

200

Carbon

Emission

sgCO2e.M

J‐1

System

Oil extraction Drying Cultivation

Rape Biodiesel ‐RED 2009/28/EC default value (cultivation & production)

Diesel ‐RED 2009/28/EC default value




Coordination Action


52

Figure 5.8: Illustrative estimates for carbon dioxide emissions from algal biomass

production in photobioreactors PBRs.

0

100

200

300

400

500

600

Jorquera(FP PBR) Jorquera(FP PBR)* Stephenson(T PBR) Stephenson(T PBR)*

Jorquera(T PBR) Jorquera(T PBR)*

Carbon

Emission

sgCO2e.M

J‐1

System

Oil extraction Drying Cultivation

Rape Biodiesel ‐RED 2009/28/EC default value (cultivation & production)

Diesel ‐RED 2009/28/EC default value

FP PBR: Flat‐plate Photobioreactor; T PBR: Tubular Photobioreactor


The impact of expanding the system boundary to include lipid production (and energy

recovery from residues) is shown in Figures 5.7 and 5.8, for raceway and PBR systems. As

might be expected, the NER for lipid production is less favourable than for biomass production

as there are losses in the conversion of the residue to heat and electricity and all the burdens

of the system are allocated to the lipid6.

Only two normalised systems, Stephenson et al (Stephenson, et al., 2010) and Kadam

(Kadam, 2001) yield more energy in the lipid than they consume. It is also apparent that

NERlipid and NERbiomass values do not vary by a constant proportion; this is due to reported

differences in the composition of the algae. For example, the NERbiomass for the Stephenson

6 The energy reported for lipid extraction in the various studies is ~0.04-0.06MJ/MJdry algal biomass. This compares with an energy demand of ~0.5-0.7MJ/MJdry algal biomass for production in a raceway pond and ~1-6MJ/MJdry algal

biomass in a PBR.


Coordination Action


53

and Lardon systems is similar, but Stephenson’s system is preferable7 for producing biodiesel

because the lipid proportion is greater. Another factor that contributes to the difference is the

oil extraction efficiency (for example, Lardon assumes ~70% oil extraction efficiency, whereas

Sander and Murthy assume ~90%).

Overall, these results indicate that producing biodiesel as the main product is likely to have

limited benefits, if assessed in terms of the system energetics. These results also illustrate the

importance of comparing systems within consistent boundaries, and it should be noted that

an alternative allocation system, for example, producing high value protein for animal feed

and allocating the impacts on the basis of market value, could change this result.

Figure 5.7: Net Energy Ratio for biomass and lipid production in raceway ponds:

comparison of normalised values.



7 I.e. it has a lower NERlipid


Coordination Action


54

Figure 5.8: Net Energy Ratio for biomass and lipid production in PBRs: comparison

of normalised values

0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

Jorquera(FP PBR)* Stephenson(T PBR)* Jorquera(T PBR)*

Net Ene

rgy R

atio

Prim

ary E

nergy Inp

ut (M

J/MJ o

utpu

t)

Net Energy Ratio in Photobioreactor

Net Energy Ratio (NER) for Oil production Net Energy Ratio (NER) for Biomass production

FP PBR: Flat‐plate Photobioreactor; T PBR: Tubular Photobioreactor


5.3 Conclusions

The LCA studies that are accessible in the literature describe a range of production systems,

but the results are difficult to compare. The normalized model presented in this chapter allows

the energetic performance of these systems to be examined on a more consistent basis and

provides some insight into the assumptions used in each of the studies. We consider that this

analysis supports the following conclusions:

• Raceway Pond Systems consistently demonstrate a lower (more desirable) NER for both

biomass and lipid production than PBR Systems

• The NER for biomass and lipid production in the normalized system described is

unattractive, or at best, marginal. This suggests that algae production may be most

attractive where energy is not the main product.

• The carbon emissions from algae biomass produced in raceway ponds is comparable to

the emissions from conventional biodiesel.

• The carbon emissions from algae biomass produced in PBRs is greater than the emissions

from conventional diesel. The principle reason for this is the electricity used for

pumping the algal broth around the system.


Coordination Action


55

• The most optimistic values for algae production in the literature come from the systems

that are the least complete. The addition of additional process steps makes the NER less

attractive in all cases.

• While the meta‐model includes some additional process steps, others might also

reasonably be included in a complete system. These include: the energy embodied in

chemical flocculant, and hexane loss during harvesting and lipid extraction. In hot

climates PBRs may also require cooling. The addition of these processes would make

the NER less attractive.

• There is a significant variation in the energy consumed in the cultivation and harvesting

phase per MJ algae (biomass or lipid) produced. Key assumptions that affect this are the

productivity of the algae, its calorific value and lipid content. (As discussed in Chapter 5,

assuming both a high productivity and high lipid content may be over optimistic.)

• Assumptions in the original studies are often obscure, or open to interpretation. For

example, the study by Kadam includes less nitrogen as an input than is contained in the

algae anticipated as an output. This may be an oversight, or the authors may have made

some additional assumption that is not explicit: it is possible that the missing nitrogen

may be recycled or come from some other source.

• Algae production requires a number of energy demanding processes. However, within the

LCA studies considered here there is no consistent hierarchy of energy consumption.

Aspects that will need to be addressed in a viable commercial system include: the

energy required for pumping, the embodied energy required for construction, the

embodied energy in fertilizer, and the energy required for drying and de‐watering.


Coordination Action


56

6. Environmental impacts of micro‐algae production

Algae production could have a wide variety of environmental impacts beyond the

consumption of energy in the production process. Impacts will vary depending on the

production technology and location, and may be positive, e.g. contributing to water

remediation, or negative e.g. emissions of hexane or fertilizer. This chapter reviews the major

environment impacts which could influence sitting decision for the cultivation of micro‐algae.

6.1 Water Resources

A reliable, low cost water supply is an important factor in the overall success of biofuel

production from micro‐algae. Per litre re of biofuel, a minimum 1.5 litres of water are required

(assuming a lipid content of 50%) (René H and Maria J, 2010). In practice, however, water use

in production systems will be much larger: fresh water needs to be added to raceway pond

systems to compensate water evaporation; water is also used for cooling Closed Systems

(PBRs).

Evaporation may be particularly high in raceway pond systems that are shallow and

mechanically mixed (Lundquist, et al., 2010). The evaporation rate may also affect the “blow

down rate (BDR)”, which is defined as the quantity of water discharged divided by the quantity

of water supplied to the pond (Lundquist, et al., 2010). In normal operation the BDR is set to

ensure that the water salinity does not exceed a certain optimal point for algae production.

One suggestion is that algae cultivation could utilize water with few competing uses, such

as seawater and brackish water from aquifers. Brackish water, however, may require pre‐

treatment if the chemical constituents of the water could inhibit algae growth. This pre‐

treatment could raise the energy demand of the process (Darzins, et al., 2010).

The distance to the water source is also an important factor in locating the cultivation site,

as 100 meters elevation could mean that 6% of the energy produced by the algae would be

used for pumping (Lundquist, et al., 2010). Consequently, coastal regions are more obviously

suitable for sustainable large scale algae production systems.

Using algae with wastewater treatment is another option for algae cultivation that has

been greatly discussed. Clarens et al. (2010) considers the different types of wastewater that

can be used, and argue that that using wastewater can greatly improve the final outcome of

the LCA. (Benemann, 2010b) argues that the final use of wastewater will be limited, and

Greenwell (2010), suggests that the best choice in the end would be seawater, as “once


Coordination Action


57

wastewater becomes the basis of a process, it is then no longer waste and will begin to have a

value”. “Waste,” he considers, “will be useful in micro‐scale operations or where remediation

is important”. An important consideration in the use of waste water is the acceptability of the

co‐products for large markets such as animal feed. In Europe, for example, the Animal Feed

Regulation bans the use of wastewater and all derived products in animal feed.

Campbell et al. (2010) and Kadam (2001) use marine water, but do not expand on the

implications of this (i.e. whether it would need desalination and what the environmental

consequences of this would be).

Stephenson et al. (2010) (and to a lesser extent Clarens et al. (2010)) are the only studies

that consider the country specific nature of the water usage and shows that it would be lower

in open systems in the UK due to higher rainfall than it would in other countries. None of the

LCA studies reviewed in the report consider the implications of using freshwater: where this

water would be sourced from, the consequences on nearby agriculture (or if it could be

recycled within the local system), and the consequences of using it in water‐scarce areas.

Although most experts agree this is the least favourable option for water supply. Neither do

any of the LCA studies consider the added‐value that could be achieved by algae helping to

treat wastewater.

6.2 Land Use

One of the suggested benefits of algae production is that it could use marginal land, and thus

would entail little additional competition on land required for food production. Land use

change has non‐obvious life cycle impacts, however, physical constrains from topography and

soil could limit the land availability for the raceway pond system: such as the installation of

large shallow ponds requires relatively flat terrain; Soil issues also have to be concerned as

their porosity/ permeability could affect the need for pond lining and sealing, which could

contribute to the environmental burdens (Lundquist, et al., 2010).

Clarens et al. (2010) shows that algae requires less land than other biofuels, while

Stephenson et al. (2010) and Campbell et al. (2010) both mention that non‐arable land can be

used, but none of the studies truly take into account the impact of this and quantitatively

shows how this compares to first generation biofuels which require cropland. This is an

important feature of algae growth, and is one of the arguments in favour of the overall

sustainability, yet is not demonstrated in the LCA studies.


Coordination Action


58

6.3 Nutrient and Fertilizer Use

Algae cultivation requires several fertilizers, primarily Nitrogen (N), Phosphorus (P) and

Potassium (K). The requirement for fertilization cannot be avoided as the algal biomass itself

consists of ~7% Nitrogen and ~1% Phosphorus. Substituting fossil fuels with algal biomass

would require a lot of fertilizer (René H and Maria J, 2010). For instance, if the EU substituted

all existing fuels with from algae biofuels this would require ~25 million tonnes of Nitrogen

and 4 million tonnes of Phosphorus. Supplying this would double the current EU capacity for

fertilizer production (van Egmond, et al., 2002).

Recycling nutrients from waste water could potentially provide some of the nutrients

required. Also, locating large algae cultivation system near waste water treatment could help it

holds the potential both for fuel production and waste water remediation.

6.4 Carbon fertilisation

Currently, large point sources of CO2 are concentrated close to major industrial and urban

areas. Research from the IPCC on Carbon Capture and Storage (CCS) identifies that, a small

proportion of large CO2 sources are close to oceans, which are potential storage locations for

CO2 and are likely to be the preferred location for algal biofuel production (Darzins, et al.,

2010). The proximity of the cultivation to a carbon source has been identified by various

experts as one of the possible limitations for growth. If we assume that a minimum of 1.8 ton

of CO2 is needed to produce 1 ton of algal biomass (Kliphuis, et al., 2010), ~1.3 billion ton of

CO2 would be required to produce the 0.4 billion m3 of biodiesel needed to supply the EU

transport market (European Environment Agency, 2008). The distance across which CO2 may

need to be transported to the cultivation site may be a concern as long CO2 transport

distances could increase the emission and energy consumption dramatically. Campbell et al.

(2010) is the only LCA study to explore the implications of this, showing that transporting

(liquefied) carbon greatly reduces the overall sustainability of the project, while using flue gas

or carbon from industry is more beneficial. It should be noted that separating CO2 from flue

gas is a highly energy consuming process (~4MJ.kg‐1) (Gambini, 2000) and so the direct use of

flue gas would be preferable. While the studies generally considered flue gas as the main

carbon source this needs further investigation as it may limit the areas of growth (a large

amount of land would have to be available near a power plant). Power plants also produce

CO2 24 hours per day, but algae will only consume it during daylight, an algal system, therefore

will only ever be able to use a small proportion of the CO2 produced from a power plant.


Coordination Action


59

The EU 2009 Renewable Energy Directive does not give any incentive to use CO2 emissions

(or wastewater), although if PBRs could capture 80% of the CO2 fed to them, they could fall

under the EU Emissions Trading Scheme (ETS), resulting in financial incentive for industry to

partake in algae production (Vernon, 2010). There is a danger, however in viewing algae as a

sequestration method as CO2 use by algal cultures is not CO2 sequestration – that comes from

algal biofuels replacing fossil fuels” (Benemann, 2010a). Campbell et al. (2010) differentiated

between the carbon released from fossil fuels (which would add new carbon to the

atmosphere) and that by biomass burning (which only re‐releases carbon previously

captured): this is an important differentiation to make.

6.5 Fossil Fuel Inputs

As discussed in the previous chapter, the majority of the of the fossil fuel inputs to algae

cultivation come from electricity consumption during cultivation sector and natural gas

consumption from the biomass drying process. Algae are quite sensitive to temperature, and

maintaining a high level of productivity could also require temperature control. If required,

both heating and cooling could require additional fossil fuel. The environmental performance

could, however, be improved by system integration that to utilize the waste heating from

power generation for drying the algal biomass.

6.6 Eutrophication

Nutrient pollution is termed eutrophication and can lead to highly undesirable changes in

ecosystem structure and function, including:

• Increased biomass of freshwater phytoplankton and periphyton

• Changes in vascular plant production, biomass, and species composition

• Reduced water clarity

• Decreases in the perceived aesthetic value of the water body

• Taste and water supply filtration problems

• Possible health risks in water supplies

• Elevated pH and dissolved oxygen depletion in the water column

• Increased probability of fish kills (Lardon, et al., 2009).


Coordination Action


60

The impact of algal aquaculture on eutrophication could be positive or negative. Negative

impacts could occur if residual nutrients in spent culture medium are allowed to leach into

local aquatic systems. On the other hand, positive impacts could occur if algae production

were to be integrated into the treatment of water bodies already suffering from

eutrophication. For example, Agricultural Research Service scientists found that 60%~90% of

nitrogen runoff and 70%~100% of phosphorus runoff can be captured from manure effluents

using an algal turf scrubber (ATS) (AquaFUELs, 2011a). Remediation efforts of polluted water

bodies suffering from algal blooms may also provide significant amounts of free waste

biomass, and this could be used for biofuel production.

In the case of macro‐algae, it has been suggested that cultivating it close to fish farms and

shrimp ponds could reduce nitrate and phosphate pollution from the farm and provide a

source of agar (Troell, et al., 1997).

6.7 Genetic Modified Algae

In the search for algae that can deliver both a high biomass productivity and a high oil content,

genetic modification is one possible option (Lundquist, et al., 2010). Applications of molecular

genetics range from speeding up the screening and selection of desirable strains, to cultivating

modified algae on a large scale. Traits that could be desirable include herbicide resistance to

prevent contamination of cultures by wild type organisms. The legal status of genetically

modified algae is somewhat unclear. In the U.S, there is question about whether applicable

laws and regulations constrain the use of Genetic Modified Algae(GMA) for biofuel

production, and there is the possibility that some companies may go ahead and use of GMA in

open systems under current regulation (or their absence) (Lundquist, et al., 2010). Open

systems present a particular challenge in terms of containment, as some culture leakage and

transfer (e.g. by waterfowl) is unavoidable. Closed bioreactors may appear more secure but

Lundquist et al., (2010) comments that PBRs are only cosmetically different from open ponds

and culture leakage is unavoidable.

In the European Union the use of genetically modified organisms is closely regulated and if

algae were genetically modified they would come under the control of European legislation

aiming to prevent the release genetically modified organisms (GMOs) and their entry into the

food chain8. There are, however, a number of projects that seek to deliver value added

8 Key legislation includes: • Directive 2001/18, as amended by Directive 2008/27/EC on the deliberate release into the environment

of genetically modified organisms;


Coordination Action


61

products through genetic engineering in the EU including the Framework Programme 7 ‐

Genetic Improvement of Algae for Value Added Products (GIAVAP) project.

6.8 Algal toxicity

Algal toxicity may be a concern where co‐products are used to produce food. Many algae

species can produce a wide variety of toxins ranging from simple ammonia to physiological

active polypeptides and polysaccharides. Effects can range from acute effects (e.g. the algae

responsible for paralytic shellfish poison may cause death) to the chronic (e.g. the toxins

produced in red tides – carrageenans – can induce carcinogenic and ulcerative tissue changes

over long periods of time). Toxin production is species and strain specific and may also

depend on environmental conditions. The presence or absence of toxins is thus difficult to

predict (Collins, 1978) (Rellán, et al., 2009).

From the perspective of producing biofuels, perhaps the most important issue is that

where co‐products are used in the human food chain producers will have to show that the

products are safe. Where algae are harvested from the wild for human consumption the

principal concern is contamination from undesirable species. From an economic perspective

algal toxins may be important and valuable products in their own right with applications in

biomedical, toxicological and chemical research.

6.9 Conclusions

Micro‐algae culture can have a diverse range of environmental impacts. Many of these

impacts are location specific, e.g. water and land use. Impacts such as the use of genetic

engineering are uncertain, but may affect what systems are viable in particular legislatures.

The impacts presented here are the ones identified as most important in the existing

literature, but should not be considered exhaustive. In any algae cultivation scheme it should

be anticipated that environmental monitoring will play an important role and will be an

ongoing requirement.

• Regulation (EC) No 1830/2003 concerning the traceability and labelling of genetically modified

organisms and the traceability of food and feed products produced from genetically modified organisms;

• Directive 2009/41/EC on the contained use of genetically modified micro-organisms. •


Coordination Action


62

7. Review of macro‐algae LCA

In the 1970’s the Energy Research and Development Authority and the American Gas

Association took over the Marine Biomass Concept programme working on cultivation and

bioconversion of the giant kelp Macrocystis pyrifera (extensive review and discussion in Bird &

Benson (1987)). The project was largely unsuccessful and research subsequently abandoned.

In 1994, Gao & McKinley then published a paper entitled ‘Use of macroalgae for marine

biomass production and CO2 remediation: a review’ (Gao and McKinley, 1994). Little research

was stimulated until recently when resurgence of interest in the potential of algal biofuels has

become intense. Thus, in spite of the relative longevity of the idea, and much current

speculation about possibilities, there are very few published studies relating to the social,

economic, environmental (including energy balances) and technical challenges of using

macroalgae as a third generation biofuel feedstock. To our knowledge no full Life‐Cycle

Analyses (LCA’s) exist in the published literature, in part this is due to lack of data with which

to populate the models.

Testimony to the fact that there is still very little available data, there are a number of recent

publications addressing algal biofuels in which macro‐algae are given scant mention, and often

only in the context of stressing the lack of information ( see for example, Chung et al. 2010;

Singh et al. 2011; John et al. 2011; Singh & Olsen 2011).

One partial LCA study on macro‐algae has been completed by Aresta et al. {, 2005 #600}.

These authors report on the development of computing software for a macro‐algal biofuel

production LCA and give some preliminary outputs of the model. They claim that the model

can be adapted to assess biofuel from either micro‐ or macroalgae and is summarised in Table

7.1. They state preliminary results of analysis for Chaetomorpha linum and Pterocladiella

capillacea grown in ponds and lipid extracted by scCO2. In the best case there was a net

energy gain in the order of 11000MJ/t dry algae, which compared to 9500MJ/t dry, gasified

microalgae. They state that distribution of separated CO2 was more energetically favourable

than the distribution of flue gas, and that the application of fresh nutrients (as opposed to

wastewater/aquaculture recycled nutrients) gave a barely positive energy balance, if not

negative.


Coordination Action


63

A recent hypothetical analysis, was carried undertaken by Goh & Lee (2010) on the macro‐

algal production capacity in Sabah, Malaysia. Along with figures for carbohydrate yields,

fermentation efficiencies and net calorific value for ethanol derived from literature, this gave a

total available energy estimate of 6.50 x 106 GJ per year. The authors go on to discuss some

challenges and constraints but in this case none of these are quantified

Table 7.1: Summary of Aresta et al. (2005) study of macro‐algae LCA

Aim To establish the energetic benefits of biofuel production from macroalgae

Assumptions • The ponds/bioreactors are situated at the coast • Nutrients are recycled from wastewater or an associated fishery • Gas transport is within the range of 100km

CO2 Capture • Considers capture from power plants ranging in size from 100‐600MW and powered by coal, oil or natural gas.

• Transport of either flue gas (if algae are resistant to NOx, SO2) or separated pure CO2

Production System Temperature, salinity, irradiance, aeration, stirring, fertilisation

Harvest Not elucidated

Drying By solar energy or recovered heat

Conversion Considers a range of processes: combustion; extraction by supercritical CO2

or organic solvents, pyrolysis, gasification, liquefaction, anaerobic

fermentation

Source: {Aresta, 2005 #600}

Despite the paucity of studies, there are a number of ongoing research projects, the outputs

of which will include LCA for macro‐algae. These are listed below.

BioMara ‐ http://www.biomara.org/

Sustainable Fuels from Marine Biomass is a joint UK and Irish Interreg IVA Project. Socio‐

economic outputs will include a microeconomic cost‐benefit analysis; a macroeconomic study

of the impacts of the development of a mari‐fuels industry in Scotland and the North of

Ireland; a techno‐economic evaluation of systems and options including biorefinery, energy

scenarios and infrastructure. Macroalgae outputs will include an Environmental Impact

Assessment (EIA); optimisation of methane yield from anaerobic digestion and scaling up;

bioethanol production and scaling up.


Coordination Action


64

Energetic Algae ‐ http://www.nweurope.eu/index.php?act=project_detail&id=4124

An Interreg IVB Project running from 2009‐2015. The project aims to establish an up to

date inventory of current and planned pilot cultivation sites with information sharing and

establishment of best practice. It will also identify political, economic, social and technical

opportunities to exploit algal biomass in North West Europe.

SuperGen ‐ http://www.supergen‐bioenergy.net/

Supergen II is the second phase of a UK based bioenergy research consortium funded by

the EPSRC. ‘Systems’ is one of the Supergen II themes and includes resource assessment;

technical, economic, environmental and social systems analyses; multi‐criteria assessment;

pathways, policies and impacts. ‘Technical performance, economic feasibility, environmental

impact and social implications of entire Bioenergy systems are to be evaluated over their full

life‐cycle.’ Marine biomass is included as a sub‐theme within the Resources theme.

7.1 Conclusions

No LCA for macro‐algae are available in the literature, although a number of research projects

are expected to publish on this subject in the near future. It is thought that LCAs will improve

the rationale behind each step of biofuel production and therefore give a key to the least

harmful and most efficient way of producing biofuel from macroalgal biomass.

The environmental impacts of macro‐algae production are also uncertain, despite

cultivation on artificial structures having been undertaken for decades now in SE Asia. Unlike

micro‐algae cultivation, water use is not considered a major obstacle. In cultivated systems the

most significant impacts are likely to be the provision of nutrients, and competition for the use

of the near‐shore area. Where macro‐algae are harvested from the wild, overharvesting may

risk damage to ecosystems that are dependent on macro‐algae as the bottom trophic level.

Specific environmental impacts include the following issues:

• Land‐Based Tank cultivation offers the advantage of possible integration of seaweed

into systems as biofilters with terrestrial aquaculture/wastewater, however land

availability may be an issue for the scale required in biofuel production.

• Kelp harvesting has been the main source of seaweed biomass in Europe so far.

Despite the resilience of kelp beds, resources are limited and the effect of harvesting

on associated biological communities is uncertain.


Coordination Action


65

• Harvesting macroalgal blooms in European coastal localities mitigates a high

environmental cost (risk of anoxia, decomposition and nutrient release, noxious

gases). The removal of large quantities of biomass and associated sediment (mostly

sand) might have some impact on coastal erosion however, latest techniques aim at

harvesting the biomass before it is stranded on beaches.


Coordination Action


66

8. Environmental impacts of macro‐algae production

Four alternative scenarios can be envisaged for harnessing macro‐algal production for

biofuels: cultivation at sea; tank‐based cultivation on land; harvest of the wild resource, and

harvest of nuisance bloom species. The potential impacts, both positive and negative, vary

between these alternatives methods; they will also be affected by scale. There is, however,

very little published data relating to the environmental impacts of cultivated macro‐algae

production for biofuels. In light of this, this chapter simply aims to summarise what little

published data is available, and give reference to other related studies that may highlight

future research needs and direction. The impacts of each cultivation scenario are considered

in terms of land and sea‐surface area required, water, fertiliser and nutrients, and ecosystem

effects.

8.1 Land use and near‐shore area use

8.1.2 Cultivation at sea

Laminaria production occurs both on land and at sea. The hatchery phase, where spores are

released, gametophytes cultivated, culture rope seeded and seedlings grown to transplant size

however requires relatively little space due to the minute nature of the early stages –

seedlings are transplanted to sea when they are between 3‐15mm. Large‐scale, commercial

production of kelps is only carried out in Asia but taking an example from Saccharina japonica

cultivation in China, 36kg of gametophytes was produced over a 3 month period (from an

initial inoculation of 0.75kg) using 100 x 20l bottles. This would be sufficient to produce 400

million sporelings, enough to seed 1300 hectares (Zhang, et al., 2008). Even at this scale it is

clear that this could be easily accommodated within the space of a normal laboratory (30‐50m

shelf space). Somewhat more space is required in order to seed the rope used to transfer the

seedlings to sea. This is hard to predict as it depends on longline set‐up, type of culture rope

used, nursery facilities etc. but based on extrapolations from pilot scale work in Ireland, a tank

of 1 cubic meter would be sufficient to produce enough seeded rope for 200m of longline, or

enough seeded string for a minimum of 1000m of longline. Over one hectare this would

require 25 cubic meter tanks for seeded rope or 5 tanks for seeded string.


Coordination Action


67

8.1.3 Tank based cultivation on land

Cultivation in tanks is more appropriate for small, fast‐growing species that need repeated

harvesting in order to maintain an appropriate stocking density, such as Ulva sp. Bruhn et al.

(2010) achieved a maximum biomass production of 45t dry weight ha‐1yr‐1 under ambient

(outside) light conditions and with minimal fertiliser input (latitude 56°N). They cite other

studies reporting energy intensive cultivation yields of 74 t dry weight ha‐1yr‐1, and non‐energy

intensive yields of 26 t dry weight ha‐1yr‐1 (Ryther et al. 1984, in (Bruhn, et al., 2010)).

Obviously yields will vary according to methods of cultivation; the figures here just serve to

give an idea of the area of land that might be required for cultivation of bioenergy significant

yields. Due to the necessity of a salt‐water supply, any tank‐based system must be coastally

located. Conflict with other land‐ or resource‐users is more likely to be in the form of the coast

as a shared landscape/leisure amenity, rather than in direct competition with agricultural or

industrial interests.

8.1.4 On‐shore facilities for sea and tank cultivation, wild harvest and bloom harvest

Land is not required on a large‐scale for macro‐algae production except in the case of land‐

based tank production. However, because of the seasonality of algae production, some

method of storage is required. The state in which the algae are stored (wet, semi‐dry or dry)

will affect the type of storage facility and space required.

All macroalgae production systems will also require on‐shore facilities in terms of offices,

warehouses and perhaps seaweed drying facilities. Harbour and pier facilities are required

with space for unloading large quantities of seaweed, as are bio‐refinery facilities, including

fermentation/digestion, separation and purification units. It is not envisaged that any of these

features will require more shore area than a normal aquaculture operation and processing

facility requires. Besides, the required land‐space does not require arable land, is not in

competition with agricultural land, and therefore the land area required for macro‐algae

cultivation can reasonably be excluded from the land for food vs energy debate.

8.2 Use of Near‐shore/Off‐shore space

Space in the near‐shore coastal environment is likely to be limited, particularly in Europe

where there is intense competition for coastal resources. Realistic projections of potential

yields (e.g. maximum 30 dry tonnes hectare‐1, based on repeated annual harvests of

Saccharina japonica in China) give some idea of the area required depending on the size of

biorefinery/biofuel installation intended.


Coordination Action


68

Further offshore space is less limited and this has been suggested as a potential solution

(e.g. (Buck and Buchholz, 2004)), but there are a number of technical, environmental and

economic questions to resolve before this becomes a reality.

8.3 Freshwater Use

Freshwater requirements for production of marine macroalgae are absolutely minimal and

limited to equipment care and maintenance.

8.4 Fertiliser and nutrients

8.4.2 Cultivation at sea:

In Europe cultivation at sea has been practiced on a very small‐scale. The use of fertilisers has

not been carried out/assessed, even experimentally ‐ all growth/biomass estimates reported

being those naturally attained. The FAO Culture of Kelp Manual states that the need for

fertiliser depends on the minimum N growth requirements of kelp and the hydrodynamic

regime of the farm area (FAO, 1989). Key considerations are the ability of the water body to

distribute nutrients, and the continual influx of non‐nutrient depleted water. The FAO states a

general requirement (for Saccharina japonica) of 20 mg NO3‐N.m‐3, but as little as 6‐10 mg

NO3‐N.m‐3 will still support production of 30 d.t.ha‐1 where there is rapid water exchange.

They give an annual fertiliser input of 2250 kg ha‐1, although what fertiliser is used, or the total

N content is not stated.

While kelps grow and survive well in nutrient poor waters (for example, exposed locations

on the west coast of Ireland support 80 wet t.ha‐1), N uptake is positively correlated with

characteristics such as heat tolerance / decay resistance and sorus (gamete) formation, and

can therefore be central to extending the growth season (and therefore the final yield of the

crop), or to producing high quality reproductive sporophytes. Maximising the growth season

has been found to dramatically enhance yield and so the importance of this point should not

be underestimated.

Every site will also be different. The Nitrates Directive / Water Framework Directive

(Directive 2000/60/EC) gives no specific protocol or criteria for fertilisation at sea although

provides a comprehensive format for monitoring of biological and physico‐chemical water

quality aimed at attaining or maintaining/improving ‘good’ status in water bodies. Criteria are

such that fertilisation in certain water bodies might not be out of the question in terms of


Coordination Action


69

compliance with the directive, but the public response would also need to be considered in

terms of acceptability.

There are other options for providing sufficient nutrients for luxuriant kelp growth that have

been proposed and in some cases experimentally, or commercially, verified. The high capacity

of kelps to remove inorganic N and P from the environment makes them potential nutrient

‘scrubbers’ from environments receiving a high nutrient loading or from those that are

eutrophic. There is quite a large body of work that has started to elucidate the capacity of

various macro‐algae to remove nutrients from wastewater sources, or from other forms of

higher trophic level aquaculture (Integrated Multi‐Trophic Aquaculture (IMTA) concept – e.g.

(Abreu, et al., 2009), or see Chopin et al. (2001) for review). While the nutrient budgets of

each aquaculture installation will be site and species specific, as an example, Sanderson (2006)

estimated that the area of S. latissima needed for remediation of a 500 tonne salmon farm in

Scotland, over the two year growth cycle, would be between 20 and 100 hectares, dependent

on the exact N incorporation into the seaweed. Based on δ15N isotope tracking, the effect of

fish farm nitrogen was detected at 500m+ (Sanderson, 2006) and up to 1350m (Handley, et al.,

2004) from the source, suggesting that a macroalgae farm of sufficient size could be located

within the area directly impacted.

Likewise, remediation of eutrophic areas resulting from land‐based nutrient inputs is a

possibility; however, while this may ameliorate an existing problem it does not provide a full

solution. Moreover, the Water Framework Directive is aiming to remove the source of the

eutrophication problems. The ideal future situation must be that there is no eutrophication

problem to remedy in the first place. Where it is deemed suitable as a mitigation method,

however, Fei (2004) identified four pre‐requisites in order that macroalgae fulfils the

mitigation role: that there is the possibility of large scale aquaculture in the eutrophic area;

that scientific and technical problems of cultivation are solved; that there are no harmful

ecological side‐effects; that cultivation is economically attractive. In Europe, all four of these

questions remain at least partially unanswered.

The feasibility of offshore macro‐algal production has been demonstrated at a concept level,

with carrier structures able to withstand offshore conditions ( >8 nautical miles from the

coast, fully exposed to all oceanographic conditions – (Buck and Buchholz, 2004) – in this case

6 m wave height and over 2 m.s‐1 current speed (Buck and Buchholz, 2004) (Buck and

Buchholz, 2005). Seaweed (Saccharina latissima) grew in fully and partially exposed locations


Coordination Action


70

although no biomass yield was reported for the fully exposed sites, and yields of 4kg

wet.weight.m‐1 was achieved at the partially exposed location. Optimisation of yield was not

the objective of this study, so, although the yield seems rather low, future improvements may

be possible. However, nutrient distribution in the offshore environment is both spatially and

temporally variable, and oceanic waters are predominantly oligotrophic so it may be that

some form of fertilisation is necessary in order to achieve economically viable yields.

Proposals include integration with offshore aquaculture, possibly itself integrated into

offshore wind farms e.g. (Buck, et al., 2008), in order that the infrastructure and

technical/maintenance costs are shared. Artificial up‐welling of nutrient rich deep water has

been suggested as a method of fertilising the relatively nutrient poor surface waters and this

has been successfully trialled in circumstances related to enhancing primary productivity e.g.

Masuda et al. (2010) (Japan), and in Norwegian fjords to enhance primary production and

thereby carrying capacity for mussel cultivation (Aure, et al., 2007). Filgueira et al. (2010) also

incorporated up‐welling into a model of aquaculture‐environment interactions and carrying

capacity. Technical and biological/ecological research into up‐welling systems is in its infancy

and the environmental consequences of this are unknown.

Another possibility is the reduction of the need for fertiliser input through strain selection.

There is the possibility of enhancing nutrient uptake characteristics by selective breeding or

careful selection of parent source populations – e.g. Macrocystis pyrifera in California

(Kopczak, et al., 1991). Successful breeding of heat‐tolerant strains allowing growth over a

longer season has also been carried out, e.g. Zhang et al. (2011).

8.4.3 Land‐based tank cultivation

As for sea cultivation, tank cultivation can also be integrated into other aquaculture systems in

order to bio‐remediate the waste‐water of fed aquaculture at the same time as producing a

valuable crop. There are many examples of this on experimental/pilot scales (e.g. Neori et al.

(2003) – Ulva lactuca with gilthead seabream, (Rodruigeza and Montaño, 2007)‐ Kappaphycus

spp. (carrageenophytes), with the milkfish Chanos chanos). In this case cultivation has a clear

environmental benefit, but care must be taken with release of cultivation water back into the

environment and continual monitoring will be necessary as nutrient rich cultivation water

released into environment could stimulate a eutrophic event.


Coordination Action


71

8.5 Macro‐algal Domestication and Genetic Engineering

Agronomically, macroalgae cultivation is in its infancy by comparison with terrestrial crops.

Domestication of kelp began in earnest in China during the mid 1950’s when the summer

sporeling method was introduced (FAO, 1989) and strain selection began to be developed in

the early 1960’s (Zhang, et al., 2011). Strain selection has led to dramatic increases in yields in

recent years and is considered vital to development of an economically viable industry. In spite

of this, there are potentially negative consequences of large‐scale release into the

environment of modified crops. Byrne & Stone (2011) state, in relation to terrestrial biofuel

crops: ‘The biosecurity risks of biomass feedstock production for bioenergy need to be

evaluated since the functional traits for bioenergy crops are also typical of weed species.’

Escape outside of the cultivation environment and establishment in the wild can lead to

establishment of environmental weeds which may impact on the ‘structural and compositional

features of biodiversity in agricultural landscapes’ (ibid.) as well as on ecosystem services. The

authors propose a risk assessment formula for assessing invasiveness, potential impact of

establishment (of either selectively bred, or genetically modified organisms), and risks of

genetic pollution for transgenic organisms. Genetic pollution relates to hybridisation between

domesticated and native crops (both between species, and within cultivars/populations). The

impacts relate to expression of hybrid vigour and introgression of foreign genes (e.g.

(Bergelson, et al., 1998) – ‘genetic engineering can substantially increase the probability of

transgene escape) or through out‐breeding depression and reduction of reproductive output.

Several authors stress that the consequences of genetic pollution are hard to evaluate and

advise extreme caution (e.g. Latham et al. (2006), also see articles in Current Opinion in

Environmental Sustainability 2011, Vol.3, pp. 1‐112). There are numerous examples of escapes

and establishment of transgenic organisms including some having long‐term environmental

and economic costs, for example the escape of Sorghum halepense, an introduced forage

grass in the USA (see Raghu et al. (2006)), and Agrostis stolonifera (creeping bentgrass), also in

the USA (see Moon et al. (2010), also for discussion of strategies of biocontainment of

transgenes for sustainable use of biotechnology). Having said this, while transgenic kelp is a

reality (see Qin et al. (2004, , 2005, , 1999), it has been limited to closed systems which the

authors stress must remain the case (Qin, et al., 2005).

Further to the above, problems common to large‐scale monocultures, particularly of non‐

native species are those of pest and disease introduction or enhancement and the risk of

transfer to wild populations. These also have the potential to affect yield and therefore should

be considered in environmental and economic projections.


Coordination Action


72

Sheppard et al. (2011) conclude: ‘We believe these future threats posed by a large new

non‐food agricultural sector driven by the bioeconomy have not been given adequate thought

or acceptance by industry and policy makers and in many cases have gone ignored.’ As the

macroalgal cultivation industry is in the very early stages, the opportunity to ensure full

consideration of unresolved issues, and incorporate knowledge from terrestrial systems need

not be missed.

8.6 Ecosystem Effects

8.6.2 Cultivation at Sea

Again there is little published data. Effects on the benthos have been demonstrated by Eklof et

al. (2005) (2006). Both these are tropical examples and relate to off‐bottom cultivation above

seagrass beds. Physical and biological changes were documented in sediment quality (became

finer), decreased sediment organic matter, decreased macroalgae and seagrass diversity,

decreased abundance and biomass of macrofauna, and a shift in seagrass community

structure – these changes, via speculated mechanisms of shading, emergence stress and

mechanical abrasion, are typically associated with a decrease in ecological quality. In the same

tropical system, Bergman et al. (2009) (2001) demonstrated changes (trophic identity,

abundance, species richness and community composition) in fish assemblages associated with

seaweed cultivation that were regarded as positive in some cases and negative in others. The

observed differences between lagoons were thought to be due to farming intensity and

substratum character.

Flow modification on a bay scale have been predicted by Grant & Bacher (2001), in Sungo

Bay, China. The bay occupies approximately 140km2 with a maximum depth of 15m and

virtually the whole bay is taken up with aquaculture of the kelp Laminaria japonica (3300

hectares in 1994) and the scallop Chlamys farreri (1333 hectares in 1994). A model of altered

current speeds and particle exchange was developed which predicted a 54% reduction in flow

and 41% reduction in particle exchange rate within the main cultivation area. The authors

state that ‘disregard for physical barriers associated with culture will result in a serious

overestimation of the particle renewal term and thus an overestimation of carrying capacity.’

They also acknowledge the need for field measurements to validate and refine the model.

Productivity of Laminaria forests is high and has been estimated, based on biomass, to be in

the range of 800‐1900 gCm‐2 (Sjotun, et al., 1995), 603‐1750 gCm‐2 (Mann, 1973). Productivity


Coordination Action


73

is actually greater than would appear based on biomass alone because of the continual

attrition of the lamina which results in a steady supply of particulate organic matter (POM), in

addition to exudation of excess fixed carbon contributing to the supply of dissolved organic

carbon (DOC) in the system. Abdullah & Fredriksen (2004) (2004) estimated total productivity

of L. hyperborea, i.e. including DOC (estimated at 26% of fixed carbon) and POM, to be as high

as 3000 gCm‐2. This large supply of organic matter subsidises trophic systems outside the kelp

ecosystem itself. For example, in South Africa, kelp derived detritus accounted for 65% of POM

in the intertidal zone – consumers in the intertidal were shown to be dependent on

productivity in the sub‐tidal and this effect was important even at sites along the coast where

there was no immediate kelp forest (Bustamante and Branch, 1996). Steneck et al. (2002) cite

further studies demonstrating carbon contributions to intertidal, offshore, shallow and deep

soft‐sediment, and terrestrial ecosystems. Large‐scale macroalgal cultivation will therefore

also contribute to the DOC and POM in the local ecosystem (although regular harvesting will

alter the overall dynamics) and this will perhaps be a positive effect. However quantification

of the contribution of organic matter and the impact on the ecosystem has not so far been

studied.

8.6.3 Land‐based Tank Cultivation

There is limited potential for effects due to the semi‐enclosed nature of the system, except for

the potential stimulation of a eutrophication event by release of nutrient enhanced water.

8.6.4 Wild Harvest

Kelp forest ecosystems have a high intrinsic ecological stability (Steneck, et al., 2002) meaning

that in themselves they are relatively resilient and can withstand considerable physical

disturbance, including that of harvesting. However, kelp individuals and ecosystems support a

very diverse and abundant flora and fauna e.g. Christie et al. (2003), Norderhaug et al. (2003)

and studies have shown that the resilience of the associated communities tends to be less

than that of kelp forest itself and the recovery times post‐harvest slower than for the kelps

(Christie, et al., 1998). Further to this, multi‐trophic interactions are largely unknown, but see

Lorentsen et al. (2010) for a comprehensive study of trophic interactions between kelps and

associated fauna and consequences of harvesting.

High productivity means that kelp ecosystems tend to provide trophic subsidies to other

inter‐related ecosystems (e.g. (Abdullah and Fredriksen, 2004) (Jørgensen and Christie, 2003))

so de‐stabilisation of kelp ecosystems has the potential to have far‐reaching consequences.


Coordination Action


74

The effect will be very much dependent on scale of the harvest. Norway has experience of

long‐term kelp harvesting with apparent stability (Vea and Ask, 2010) but a recent paper

suggests that the effects of the harvest (in terms of local and offshore ecosystems) are not

fully appreciated at present and may be more detrimental than previously thought (Lorentsen,

et al., 2010).

For a full discussion of the potential effects of harvesting see Ratcliff (Ratcliff J (in press),

2010).

8.6.5 Harvest of blooms

Liu et al. (2009) found large‐scale porphyra aquaculture in combination with a eutrophic water

body stimulated a massive (400km2) Enteromorpha prolifera bloom in China. Removal of

bloom biomass is important to prevent the formation of the anoxic conditions that follow

during the degradation of the biomass, which can be extremely detrimental to other

organisms. Bloom events resulting from eutrophic water bodies are not uncommon and can

result in substantial quantities of biomass however, they may be unpredictable in terms of

time and space which results in logistical difficulties for harvesting and processing, particularly

in terms of transport of a wet, bulky material, which will impact the GHG emissions budget for

the process. Further to this, as mentioned above, the ideal is that the eutrophication initially

stimulating the bloom is dealt with in the first place rather than attempting to mitigate the

effects of a different problem.

8.7 Environmental Contamination

During the nursery stage of cultivation antibiotics (chloromycetin) and germanium dioxide are

sometimes used in order to control contamination of the gametophyte and juvenile

sporophyte cultures with alginic acid decomposing bacteria and diatoms (respectively). In

terms of contamination, standard laboratory procedures avoiding release of these agents to

the environment should be sufficient to avoid potential negative environmental

consequences.

Solid waste will be mainly in the form of ropes, buoys and structural components of the

farm. Longevity is of obvious importance from an economic perspective however, particularly

in terms of rope requirements, which over a large‐scale farm will amount to a substantial

quantity, consideration of materials will be important. In China, seeding has predominantly

been carried out on palm‐fibre rope which is biodegradable. However, in Europe, trial


Coordination Action


75

cultivation projects have tended to use synthetics – either polypropylene or polyvinyl alcohol

based vinylon fibre. Potential for re‐use or recycling should be assessed.

Airborne emissions during both production (above the macro‐algal crop) and storage have

not been measured. However, kelps are known to emit volatile short‐lived organic compounds

like organo‐iodines and molecular iodine which play a major role in the iodine biogeochemical

cycle. The emissions are significant especially when kelps are subjected to stress such as

emersion, exposure to ultraviolet radiation or elevated ozone. These volatile organic

compounds could contribute to marine aerosol formation and have a direct effect on the

earth’s climate (see Carpenter et.al. (2000)). Emissions of H2S during wet storage would also

need to be quantified and monitored.

8.8 Conclusions

The environmental impact of macro‐algae production for biofuels has yet to be assessed.

Nevertheless, there is evidence that this source of biomass is associated with a range of

impacts. In cultivated systems the most significant impacts are likely to be the provision of

nutrients, and competition for the use of the near‐shore area. Where macro‐algae are

harvested from the wild, overharvesting may risk damage to ecosystems that are dependent

on macro‐algae.

9. Conclusions and recommendations

This report examines the available literature on biofuel production from micro‐algae and

macro‐algae, focusing on the life cycle and environmental impacts reported, and the

methodologies that have been used to assess them. This examination was supplemented with

information gathered from experts and stakeholders.

9.1 Conclusions for micro‐algae LCA, and environmental impacts

The impacts of micro‐algal biofuel production have been assessed using a variety of life

cycle assessment (LCA) approaches. These studies share a common aspiration to identify

production bottlenecks and help steer the future development of algae biofuel technology.

Clarens (2010b), for example, claims that one of the strengths of LCA is that it “can provide a

quantitative road map, and when it is combined with financial modelling can provide an

excellent barometer for how to develop algae bioenergy”. Yet, the extent to which the studies

meet this aspiration appears to be somewhat limited.


Coordination Action


76

Issues of concern include:

• The conceptual, and often incomplete, nature of the systems under investigation, and

the absence of coherent and well designed processes.

• The limited sources of primary data upon which process assumptions are based, and

the extrapolation of laboratory data to production scale. The transparency of

assumptions is also poor.

• The validity of specific assumptions, particularly those relating to the biomass and

lipid productivity, has been called into question, and may be over optimistic.

• The use of inconsistent boundaries, functional units and allocation methodologies

impedes comparison between studies.

Despite these shortcomings, and the concerns voiced by stakeholders about the extent to

which the existing LCA can be considered representative, this examination of LCA studies

suggest that:

• The net energy ratio (NER) for biomass or lipid production in a simplistic but

normalized system is unattractive, or, at best, marginal. This suggests that algae

production may be most attractive where energy is not the main product.

• Carbon emissions from algae biomass – calculated using default emission factors –

produced in raceway ponds are comparable to the emissions associated with

conventional biodiesel. The carbon emissions from algae biomass produced in PBRs

are greater than the emissions associated with conventional diesel. The principle

reason for this is the electricity used to pump the algal broth around the system.

• Raceway Pond Systems consistently demonstrate a more attractive energy ratio than

PBR Systems (it should also be borne in mind that a commercial system may

combine elements of both).

• Algae production requires a number of energy demanding processes. However, within

the LCA studies considered here there is no consistent hierarchy of energy

consumption. Aspects that will need to be addressed in a viable commercial system

include: the energy required for pumping, the embodied energy required for

construction, the embodied energy in fertilizer, and the energy required for drying

and de‐watering.


Coordination Action


77

An important caveat to these conclusions is that they reflect the state of the existing academic

literature, and this is inevitably an imperfect reflection of systems being evaluated by

companies. It is quite possible that many of the challenges identified have been addressed,

but that the information about how this has been achieve is yet to make it into the public

domain.

In the absence of commercially operating systems, attempts to assess the lifecycle and

environmental impacts of algae production are necessarily based on assumed process

descriptions. This does not negate the value of these studies, but it does mitigate in favour of

a cautious approach to their interpretation. Future LCA are can be improved if these issues

identified above are addressed, but perhaps the greatest opportunity for improvement lies

obtaining good quality primary data from pilot scale systems and improving the extrapolation

of this data to full scale systems.

Looking at the environmental impacts of algae production, it is apparent that micro‐algae

culture can have a diverse range of environmental impacts, but, that depending on how the

system is configured, these may be positive or negative. Other than the energy balance,

possibly the most important environmental aspect of micro‐algae culture that needs to be

considered is water management: both the water consumed by the process, and the

emissions to water courses from the process.

9.2 Conclusions for macro‐algae LCA and environmental impacts

No LCA for macro‐algae are available in the literature, although a number of research projects

are expected to publish on this subject in the near future.

The environmental impacts of macro‐algae production are also uncertain. Unlike micro‐

algae cultivation, water use is not considered a major obstacle. In cultivated systems the most

significant impacts are likely to be the provision of nutrients, and competition for the use of

the near‐shore area. Where macro‐algae are harvested from the wild, overharvesting may risk

damage to ecosystems that are dependent on macro‐algae as the bottom trophic level.


Coordination Action


78

10. References

ABDULLAH, M. and FREDRIKSEN, S. (2004) Production, respiration and exudation of dissolved organic matter by the kelp Laminaria hyperborea along the west coast of Norway. Journal of the Marine Biological Association of the UK, 84, 887‐891.

ABO (2010) Algal Biomass Organization Questions Accuracy of University of Virginia Algae Life Cycle Study [Internet], Algal Biomass Organisation (ABO), available from: <http://www.algalbiomass.org/media‐center/news‐press‐releases/abo‐press‐releases/algal‐biomass‐organization‐questions‐accuracy‐of‐university‐of‐virginia‐algae‐life‐cycle‐study‐2/> [Accessed: 13 May 2011].

ABREU, M., VARELA, D., HENRÍQUEZ, L., VILLARROEL, A., YARISH, C., SOUSA‐PINTO, I. and BUSCHMANN, A. (2009) Traditional versus Integrated Multi‐Trophic Aquaculture of Gracilaria chilensis C.J. Bird, J. McLachlan & E.C. Oliveira: Productivity and physiological performance. Aquaculture 293: 211‐220. .

AMIN, S. (2009) Review on biofuel oil and gas production processes from microalgae. Energy Conversion and Management, 50, 1834‐1840.

APT, K. E. and BEHRENS, P. W. (1999) COMMERCIAL DEVELOPMENTS IN MICROALGAL BIOTECHNOLOGY. Journal of Phycology, 35, 215‐226.

AQUAFUELS (2010) Project Info [Internet], available from: <http://www.aquafuels.eu/project‐info.html> [Accessed: 16/7/2010,

AQUAFUELS (2011a) Report on Biology and Biotechnology of algae with indication of criteria for strain selection.

AQUAFUELS (2011b) Task 1.9 Mapping of natural resources in artificial, marine and freshwater bodies.

AURE, J., STRAND, Ø., ERGA, S. and STROHMEIER, T. (2007) Primary production enhancement by artificial upwelling in a western Norwegian fjord. Marine Ecology Progress Series 352, 39‐52.

BARSANI, L. and GUALTIERI, P. (2006) Algae Anatomy, Biochemistry, and Biotechnology. Boca Raton, FL. CRC Press.

BENEMANN, J. R. (2010a) Microalgae Biofuels and Animal Feeds: An Introduction.

BENEMANN, J. R. (2010b) Questionnaire about LCA on Microalgae.

http://www.algalbiomass.org/media-center/news-press-releases/abo-press-releases/algal-biomass-organization-questions-accuracy-of-university-of-virginia-algae-life-cycle-study-2/�

http://www.algalbiomass.org/media-center/news-press-releases/abo-press-releases/algal-biomass-organization-questions-accuracy-of-university-of-virginia-algae-life-cycle-study-2/�

http://www.aquafuels.eu/project-info.html�

http://www.aquafuels.eu/project-info.html�


Coordination Action


79

BENEMANN, J. R. and OSWALD, W. J. (1996) Systems and economic analysis of microalgae ponds for conversion of CO2 to biomass. Department of Energy, Pittsburgh Energy Technology Center.

BERGELSON, J., PURRINGTON, C. and WICHMANN, G. (1998) Promiscuity in transgenic plants. Nature, 395.

BERGMAN, K., SVENSSON, S. and ÖHMAN, M. (2009) Influence of algal farming on fish assemblages. Marine Pollution Bulletin, 42 1379‐1389.

BIRD, K. and BENSON, P. (1987) Seaweed Cultivation for Renewable Resources., Elsevier Science Publishers.

BOGUSKI, T. K., HUNT, R. G., CHOLAKIS, J. M. and FRANKLIN, W. E. (1996) Environmental Life‐Cycle Assessment. New York, London, McGraw‐Hill.

BRENNAN, L. and OWENDE, P. (2010) Biofuels from microalgae—A review of technologies for production, processing, and extractions of biofuels and co‐products. Renewable and Sustainable Energy Reviews, 14, 557‐577.

BRUHN, A., DAHL, J., NIELSEN, H., NIKOLAISEN, L., RASMUSSEN, M., MARKAGER, S., OLESEN, B., ARIAS, C. and JENSEN, P. (2010) Bioenergy potential of Ulva lactuca: biomass yield, methane production and combustion. Bioresource Technology.

BUCK, B. and BUCHHOLZ, C. (2004) The offshore‐ring: A new system design for the open ocean aquaculture of macroalgae. Journal of Applied Phycology, 16, 355‐368.

BUCK, B. and BUCHHOLZ, C. (2005) Response of offshore cultivated Laminaria saccharina to hydrodynamic forcing in the North Sea. Aquaculture, 250, 674‐691.

BUCK, B., KRAUSE, G., MICHLER‐CIELUCH, T., BRENNER, M., BUCHHOLZ, C., BUSCH, J., FISCH, R., GEISEN, M. and ZIELINSKI, O. (2008) Meeting the quest for spatial efficiency: progress and prospects of extensive aquaculture within offshore wind farms. Helgoland Marine Research, 62, 269‐281.

BUSTAMANTE, H. and BRANCH, G. (1996) The dependence of intertidal consumers on kelp‐derived organic matter on the west coast of South Africa. Journal of Experimental Marine Biology and Ecology, 196, 1‐28.

BYRNE, M. and STONE, L. (2011) The need for ‘duty of care’ when introducing new crops for sustainable agriculture. Current Opinion in Environmental Sustainability, 3, 50‐54.

CARLSSON, A. S., VAN BEILIN, J. B., MOLLER, R. and CLAYTON, D. (2007) Micro‐ and Macro‐algae: Utility for Industrial Applications., University of York, CPL Press.


Coordination Action


80

CARPENTER, L., MALIN, G., LISS, P. and KUPPER, F. (2000) Novel biogenic iodine‐containing trihalomethanes and other short‐lived halocarbons in the coastal East Atlantic. Global Biogeochemical Cycles, 12, 1191‐1204.

CHEN, F. (1996) High cell density culture of microalgae in heterotrophic growth. Trends in Biotechnology, 14, 421‐426.

CHEN, F. and JOHNS, M. (1995) A strategy for high cell density culture of heterotrophic microalgae with inhibitory substrates. Journal of Applied Phycology, 7, 43‐46.

CHEN, F. and JOHNS, M. R. (1996) Heterotrophic growth of Chlamydomonas reinhardtii on acetate in chemostat culture. Process Biochemistry, 31, 601‐604.

CHISTI, Y. (2007) Biodiesel from microalgae. Biotechnology Advances, 25, 294‐306.

CHISTI, Y. (2008a) Biodiesel from microalgae beats bioethanol. Trends in Biotechnology, 26, 126‐131.

CHOPIN, T., BUSCHMANN, A., HALLING, C., TROELL, M., KAUTSKY, N., NEORI, A., KRAEMER, G., ZERTUCHE‐GONZÁLEZ, J., YARISH, C. and NEEFUS, C. (2001) Minireview. Integrating seaweeds into marine aquaculture systems: A key toward sustainability. Journal of Phycology, 37, 975‐986.

CHRISTIE, H., FREDRIKSEN, S. and RINDE, E. (1998) Regrowth of kelp and colonization of epiphyte and fauna community after kelp trawling at the coast of Norway. Hydrobiologia, 375/376, 49‐58.

CHRISTIE, H., JORGENSEN, N., NORDERHAUG, K. and WAAGE‐NIELSEN, E. (2003) Species distribution and habitat exploitation of fauna associated with kelp (Laminaria hyperborea) along the Norwegian coast. Journal of the Marine Biological Association of the UK, 83, 687‐699.

CLARENS, A., RESURRECCION, E., WHITE, M. and COLOSI, L. (2010) Environmental Life Cycle Comparison of Algae to Other Bioenergy Feedstocks. Environmental Science Technology, 44, 1813‐1819.

COLLINS, M. (1978) Algal Toxins. Microbiological reviews, 42, 726‐746.

DARZINS, A., PIENKOS, P. and EDYE, L. (2010) Current Status and Potential for Algae Biofuels Production. [Available from: <Error! Hyperlink reference not valid..

DE SMET, B., WHITE, P. R. and OWENS, J. W. (1996) Environmental life‐cycle assessment. New York, McGraw‐Hill.


Coordination Action


81

DEMIRBAS, A. (2009) Progress and recent trends in biodiesel fuels. Energy Conversion and Management, 50, 14‐34.

EC (2009) Directive on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC. IN E. COMMISSION (Ed.). Brussels, Belgium, European Commission.

EC (2010) ILCD Handbook: General guide for Life Cycle Assessment ‐ Detailed guidance. Joint Research Centre, Institute for Environment and Sustainability,European Commission.

EKLÖF, J., DE LA TORRE CASTRO, M., ADELSKÖLD, L., JIDDAWI, N. and KAUTSKY, N. (2005) Differences in macrofaunal and seagrass assemblages in seagrass beds with and without seaweed farms. Estuarine, Coastal and Shelf Science, 63, 385‐396.

EKLÖF, J., HENRIKSSON, R. and KAUTSKY, N. (2006) Effects of tropical open‐water seaweed farmin on seagrass ecosystem structure and function. Marine Ecology Progress Series 352, 73‐84.

EUROPEAN ENVIRONMENT AGENCY (2008) Greenhouse Gas Emission Trends and Projections in Europe 2008. European Environment Agency, Copenhagen, Denmark.

FAO (1989) Culture of kelp (Laminaria japonica) in China. Training manual 89/5. Food and Agriculture Organization of the United Nations (FAO). [Available from: < http://www.fao.org/docrep/field/003/AB724E/AB724E00.htm >].

FEI, X. (2004) Solving the coastal eutrophication problem by large scale seaweed cultivation. Hydrobiologia, 512, 145‐151.

FILGUEIRA, R., GRANT, J., STRAND, Ø., ASPLIN, L. and AURE, J. (2010) A simulation model of carrying capacity for mussel culture in a Norwegian fjord: Role of induced upwelling. Aquaculture, 308, 20‐27.

FONSECA, D. (August 2010) Questionnaire about LCA on Microalgae.

GAO, K. and MCKINLEY, K. (1994) Use of macroalgae for marine biomass production and CO2 remediation: a review. Journal of Applied Phycology, 6, 45‐60.

GNANSOUNOU, E., DAURIAT, A., VILLEGAS, J. and PANICHELLI, L. (2009) Life cycle assessment of biofuels: Energy and greenhouse gas balances. Bioresource Technology, 100, 4919‐4930.

GRAHAM, L. E. and WILCOX, L. W. (2000) Algae. Upper Saddle River, NJ, Prentice Hall Inc.

GRANT, J. and BACHER, C. (2001) A numerical model of flow modification induced by suspended aquaculture in a Chinese bay. Canadian Journal of Fisheries and Aquatic Sciences, 58, 1003‐1011.

http://www.fao.org/docrep/field/003/AB724E/AB724E00.htm�


Coordination Action


82

GRAVERHOLT, O. and ERIKSEN, N. (2007) Heterotrophic high‐cell‐density fed‐batch and continuous‐flow cultures of <i>Galdieria sulphuraria</i> and production of phycocyanin. Applied Microbiology and Biotechnology, 77, 69‐75.

HANDLEY, L., DAVIES, I., ROBINSON, C., RAVEN, J. and HADFIELD, B. (2004) Fish‐farm nitrogen: It’s area of biological impact (a scoping study). Final Project Report to The Crown Estate. [Available from: <www.thecrownestate.co.uk/aqua_fish_farm_nitrogen.pdf>].

HASE, R., OIKAWA, H., SASAO, C., MORITA, M. & WATANABE, Y. (2000) Photosynthetic production of microalgal biomass in a raceway system under greenhouse conditions in Sendai city. Journal of Bioscience and Bioengineering, 89, 157‐163.

HEIJUNGS, R., GUINEE, J. B., HUPPES, G., LANKREIJER, R. M., UDO DE HAES, H. A., WEGENER SLEESWIJK, A., ANSEMS, A. M. M., EGGELS, P. G., VAN DUIN, R. and DE GOEDE, H. P. (1992) Environmental life cycle assessment of products: Guide LCA. Centre of Environmental Science (CML), Unilever, Leiden, The Netherlands.

IEA, B. (2010) Algae – The Future for Bioenergy? Summary and conclusions from the IEA Bioenergy ExCo64 Workshop.

ISO (1997) Environmental management‐life cycle assessment‐principle and framework. ISO.

JØRGENSEN, N. and CHRISTIE, H. (2003) Diurnal, horizontal and vertical dispersal of kelp‐associated fauna. Hydrobiologia 503, 69‐76.

JORQUERA, O., KIPERSTOK, A., SALES, E. A., EMBIRUÇU, M. and GHIRARDI, M. L. (2010) Comparative energy life‐cycle analyses of microalgal biomass production in open ponds and photobioreactors. Bioresource Technology, 101, 1406‐1413.

KADAM, K. L. (2001) Microalgae production from power plant flue gas: Environmental implications on a life cycle basis. National Renewable Energy Laboratory.

KLIPHUIS, A. M. J., DE WINTER, L., VEJRAZKA, C., MARTENS, D. E., JANSSEN, M. and WIJFFELS, R. H. (2010) Photosynthetic efficiency of Chlorella sorokiniana in a turbulently mixed short light‐path photobioreactor. Biotechnology Progress, 26, 687‐696.

KOPCZAK, C., ZIMMERMAN, R. and KREMER, J. (1991) Variation in nitrogen physiology and growth among geographically isolated populations of the giant kelp Macrocystis pyrifera (Phaeophyta). Journal of Phycology, 27, 149‐158.

LARDON, L., HELIAS, A., SIALVE, B., STEYER, J. P. and BERNARD, O. (2009) Life‐Cycle Assessment of Biodiesel Production from Microalgae. Environmental Science & Technology, 43, 6475‐6481.

http://www.thecrownestate.co.uk/aqua_fish_farm_nitrogen.pdf%3e%5d�


Coordination Action


83

LATHAM, J., WILSON, A. and STEINBRECHER, R. (2006) The Mutational Consequences of Plant Transformation. Journal of Biomedicine and Biotechnology. 1‐7.

LIU, D., KEESING, J., XING, Q. and SHI, P. (2009) World’s largest macroalgae bloom caused by expansion of seaweed aquaculture in China. Marine Pollution Bulletin 58, 888‐895.

LORENTSEN, S.‐H., SJOTUN, K. and GRÉMILLET, D. (2010) Multi‐trophic consequences of kelp harvest. . Biological Conservation, 143, 2054‐2062.

LUNDQUIST, T. J., WOERTZ, I. C., QUINN, N. W. T. and BENEMANN, J. R. (2010) A Realistic Technology and Engineering Assessment of Algae Biofuel Production. Energy Biosciences Institute, University of California, Berkeley, California.

MACADAM, J. (2010) Combined Heat & Power and District Heating:the urban context. A lecture to the MSc Course in Sustainable Energy Futures, Imperial College London. London.

MANN, K. (1973) Seaweeds: Their productivity and strategy for growth. Science 182, 975‐981.

MASUDA, T., FURUYA, K., KOHASHI, N., SATO, M., TAKEDA, S., UCHIYAMA, M., HORIMOTO, N. and ISHIMARU, T. (2010) Lagrangian observation of phytoplankton dynamics at an artificially enriched subsurface water in Sagami Bay, Japan. Journal of Oceanography, 66, 801‐813.

METEOTEST database Meteonorm.

MOLINA GRIMA, E., BELARBI, E. H., ACIEN FERNANDEZ, F. G., ROBLES MEDINA, A. and CHISTI, Y. (2003) Recovery of microalgal biomass and metabolites: process options and economics. Biotechnology Advances, 20, 491‐515.

MOON, H., ABERCROMBIE, J., KAUSCH, A. and STEWART, J. C. (2010) Sustainable Use of Biotechnology for Bioenergy Feedstocks. Environmental Management, 46, 531‐538.

NECTON (1990) Estudo de localização de unidade industrial para produção de microalgas em portugal.

NEORI, A., MSUYA, F., SHAULI, L., SCHUENHOFF, A., KOPEL, F. and SHPIGEL, M. (2003) A novel three‐stage seaweed (Ulva lactuca) biofilter design for integrated mariculture. Journal of Applied Phycology, 15, 543‐553.

NORDERHAUG, K., FREDRIKSEN, S. and NYGAARD, K. (2003) Trophic importance of Laminaria hyperborea to kelp forest consumers and the importance of bacterial degradation to food quality. Marine Ecology Progress Series 255, 135‐144.

OILGAE (2010) Algae Cultivation.


Coordination Action


84

PATIL, V., TRAN, K. Q. and GISELRØD, H. R. (2008) Towards sustainable production of biofuels from microalgae. International Journal of Molecular Sciences, 9, 1188‐1195.

PULZ, O. (2001a) Photobioreactors: production systems for phototrophic microorganisms. Applied Microbiology and Biotechnology, 57, 287‐293.

PULZ, O. (2001b) Photobioreactors: production systems for phototrophic microorganisms. Appl Microbiol Biotech, 57.

QIN, S., JIANG, P. and TSENG, C. (2004) Molecular biotechnology of marine algae in China. Hydrobiologia 512, 21‐26.

QIN, S., JIANG, P. and TSENG, C. (2005) Transforming kelp into a marine bioreactor. Trends in Biotechnology, 23, 264‐268.

QIN, S., SUN, G.‐Q., JINAG, P., ZOU, L.‐H., WU, Y. and TSENG, C.‐K. (1999) Review of genetic engineering of Laminaria japonica (Laminariales, Phaeophyta) in China. Hydrobiologia 398/399, 469‐472.

RAGHU, S., ANDERSON, R., DAEHLER, C., DAVIS, A., WIEDENMANN, R., SIMBERLOFF, D. and MACK, R. (2006) Adding Biofuels to the Invasive Species Fire? Science 313, 1742.

RATCLIFF J (IN PRESS) ( 2010) Evaluation of the existing and future biomass potential of the wild kelp resource in the UK and Ireland. Proceedings of Bioten ‐ Engaging UK Bioenergy.

RAYNOLDS, M., FRASER, R. and CHECKEL, D. (2000a) The relative mass‐energy‐economic (RMEE) method for system boundary selection Part 1: A means to systematically and quantitatively select LCA boundaries. The International Journal of Life Cycle Assessment, 5, 37‐46.

RAYNOLDS, M., FRASER, R. and CHECKEL, D. (2000b) The relative mass‐energy‐economic (RMEE) method for system boundary selection. Part 2: Selecting the Boundary Cut‐off Parameter (ZRMEE) and its Relationship to Overall Uncertainty. The International Journal of Life Cycle Assessment, 5, 96‐104.

RELLÁN, S., OSSWALD, J., SAKER, M., GAGO‐MARTINEZ, A. and VASCONCELOS, V. (2009) First detection of anatoxin‐a in human and animal dietary supplements containing cyanobacteria. Food and Chemical Toxicology, 47, 2189‐2195.

RENÉ H, W. and MARIA J, B. (2010) An Outlook on Microalgal Biofuels. Science, 329, 796‐799.

RODRUIGEZA, M. and MONTAÑO, M. (2007) Bioremediation potential of three carrageenophytes cultivated in tanks with seawater from fish farms. Journal of Applied Phycology, 19, 755‐762.


Coordination Action


85

SAIC (2006) Life Cycle Assessment: Principles and Practice. Scientific Applications International Corporation (SAIC). National Risk Management Research Laboratory, U.S. Environmental Protection Agency, Cincinnati, USA.

SANDER, K. and MURTHY, G. S. (2010) Life cycle analysis of algae biodiesel. The International Journal of Life Cycle Assessment, 1‐11.

SANDERSON, J. (2006) Reducing the environmental impact of seacage fish farming through cultivation of seaweed. Scottish Association of Marine Science,

SCHENK, M. P., THOMAS‐HALL, S. R., STEPHENS, E., MARX, U. C., MUSSGNUG, J. H., POSTEN, C., KRUSE, O. and HANKAMER, B. (2008) Second Generation Biofuels: High‐Efficiency Microalgae for Biodiesel Production. Bioenergy Research, 1, 20‐43.

SETAC (1993) Guidelines for Life‐Cycle Assessment: a ‘Code of Practice' from the workshop held at Sesimbra, Portugal, 31 March ‐ 3 April 1993 Society of Environmental Toxicology and Chemistry (SETAC) Journal Environmental Science and Pollution Research

SHEEHAN, J., CAMOBRECO, V., DUFFIELD, J., GRABOSKI, M. and SHAPOURI, H. (1998a) Life cycle inventory of biodiesel and petroleum diesel for use in an urban bus. Golden, CO (US).

SHEEHAN, J., DUNAHAY, T., BENEMANN, J. & ROESSLER, P. (1998b) A Look Back at the US Department of Energy’s Aquatic Species Program—Biodiesel from Algae. U. D. O. ENERGY., Golden, CO.

SHEPPARD, A., GILLESPIE, I., HIRSCH, M. and BEGLEY, C. (2011) Biosecurity and sustainability within the growing global bioeconomy. Current Opinion in Environmental Sustainability, 3, 4‐10.

SIALVE, B., BERNET, N. and BERNARD, O. (2009) Anaerobic digestion of microalgae as a necessary step to make microalgal biodiesel sustainable. Biotechnology Advances, 27, 409‐416.

SJOTUN, K., FREDRIKSEN, S., RUENESS, J. and LEIN, T. (1995) Ecological studies of the kelp Laminaria hyperborea (Gunnerus) Foslie in Norway In Ecology of Fjords and Coastal Waters., Elsevier Science.

STENECK, R., GRAHAM, M., BOURQUE, B., CORBETT, D., ERLANDSON, J., ESTES, J. and TEGNER, M. (2002) Kelp forest ecosystems: biodiversity, stability, resilience and future. Environmental Conservation 29, 436‐459.

STEPHENSON, A. L., KAZAMIA, E., DENNIS, J. S., HOWE, C. J., SCOTT, S. A. and SMITH, A. G. (2010) Life‐Cycle Assessment of Potential Algal Biodiesel Production in the United Kingdom: A Comparison of Raceways and Air‐Lift Tubular Bioreactors. Energy & Fuels, 100‐112.


Coordination Action


86

TREDICI, M. R. (2010) Photobiology of microalgae mass cultures: understanding the tools for the next green revolution. Biofuels, 1, 143‐162.

TROELL, M., HALLING, C., NILSSON, A., BUSCHMANN, A. H., KAUTSKY, N. and KAUTSKY, L. (1997) Integrated marine cultivation of Gracilaria chilensis (Gracilariales, Rhodophyta) and salmon cages for reduced environmental impact and increased economic output. Aquaculture, 156, 45‐61.

VAN EGMOND, K., BRESSER, T. and BOUWMAN, L. (2002) The European Nitrogen Case. AMBIO: A Journal of the Human Environment, 31, 72‐78.

VEA, J. and ASK, E. (2010) Creating a sustainable commercial harvest of Laminaria hyperborea, in Norway. Journal of Applied Phycology.

VIEIRA, V. V. (2010) Personal communication about LCA on Microalgae.

WANG, B., LI, Y., WU, N. and LAN, C. Q. (2008) CO2 bio‐mitigation using microalgae. Applied Microbiology and Biotechnology, 79, 707‐718.

XU, H., MIAO, X. and WU, Q. (2006) High quality biodiesel production from a microalga Chlorella protothecoides by heterotrophic growth in fermenters. Journal of Biotechnology, 126, 499‐507.

ZHANG, J., LIU, Y., YU, D., SONG, H., CUI, J. and LIU, T. (2011) Study on high‐temperature‐resistant and high‐yield Laminaria variety “Rongfu”. Journal of Applied Phycology.

ZHANG, Q., QU, S., CONG, Y., LUO, S. and TANG, X. (2008) High throughput culture and gametogenesis induction of Laminaria japonica gametophyte clones. Journal of Applied Phycology, 20, 205‐211.


Coordination Action


87

11. Annex 1: Heterotrophic Microalgae

Heterotrophic microalgae are grown in closed culture systems that are equivalent to

conventional fermentation technologies (Apt and Behrens, 1999). In these systems,

heterotrophic algae use organic Carbon as their sole source of energy and Carbon. This

removes the issue of light limitation that dominates most aspects of photobioreactor and

raceway pond design. However since heterotrophic microalgae are grown on organic

material that could itself be used for biofuel production, it is incorrect to view this as a

primary biomass production system. Heterotrophic cultivation of microalgae should be

viewed as an alternative technology platform for conversion of biomass into biofuels, which

can produce biodiesel from carbohydrate feedstocks. The fundamental design of

heterotrophic bioreactors consists of a closed vessel that contains algae and their nutrients in

optimal conditions to maximise productivity. Vessels can range in size from 1‐500,000 litres

and their shape is dictated by economic and structural factors, unlike photobioreactors which

must maximise their surface area for the penetration of light (Apt and Behrens, 1999). The

culture medium is similar to that used for photobioreactors, with nutrients such as Nitrogen

and Phosphorous added in the correct proportions for the specific algal strain. However

heterotrophic bioreactors also introduce carbohydrates into the culture medium. Typical

carbohydrates that are used as a substrate for heterotrophic microalgal growth can be

glucose, acetate or corn powder hydrosylate (Xu, et al., 2006). Sterile air is introduced into

the culture medium at high pressure and flow rate to ensure appropriate gas exchange and

optimal conditions are maintained by constant monitoring and control of temperature, pH

and dissolved O2 and CO2 within the culture medium (Apt and Behrens, 1999).

It is possible to cultivate heterotrophic microalgae in either batch‐based or continuous

culture systems. Batch based systems provide a fixed amount of culture medium and organic

substrate to grow a batch of microalgae, which are completely harvested after a fixed period.

The major challenge of batch cultivation is to provide sufficient supply of nutrients and

substrate to ensure high cell density, without exceeding concentrations that inhibit algal

growth (Chen, 1996). This issue can be addressed using fed‐batch cultivation, which introduces

culture medium and substrate in response to demand signals such as increased oxygen

tension (Graverholt and Eriksen, 2007). Continuous culture of heterotrophic algae is usually


Coordination Action


88

performed in a chemostat, where fresh medium is introduced at the same rate as the culture

is harvested (Chen and Johns, 1996). This maintains the culture volume and cell concentration

at a steady state to optimise productivity. Alternative continuous culture systems include

membrane bioreactors. These are essentially the same as chemostat cultures apart from the

addition of a selectively permeable membrane that retains algae in the main vessel while

allowing some culture medium to wash through (Chen and Johns, 1995). This prevents

inhibitory compounds such as sodium from accumulating and reducing the stability of the

culture (Chen, 1996).


Coordination Action


89

12. Annex 2: Review of existing micro algae LCA Studies

12.1 Kadam 2001/2

Kadam (2001; 2002) compared a a conventional coal‐fired power station with one in which

coal was co‐fired with algae cultivated from recycling power plant flue gas as a source of CO2.

The system is based in the southern USA, where there is a high incidence of solar radiation.

12.1.2 Functional Unit and System Boundaries

The functional unit considered was the quantity of coal and algae necessary to produce 1

MW of electricity. The system boundaries are shown in Figure 10.1

Figure 10.1: General system boundaries for the comparison of electricity production

via coal firing vs. coal/algae co‐firing in Kadam’s study

Source: (Kadam, 2002)

12.1.3 Source of Data

Primary data was used where possible for algae cultivation, while the US EPA and literature

data was used for the other processes. According to Kadam’s own analysis, the data was


Coordination Action


90

reliable and complete at the time of writing. The system was run using TEAM® (v 3.0)

software.

12.1.4 Process

The algae cultivation was assumed to take up 1000 ha, using approximately 60% of the

emitted CO2 of a 50MW plant. Two potential pathways were investigated: direct injection of

the flue gas, and monoethanolamine (MEA) extraction. The former process includes

compression, dehydration and transportation of the CO2 to the algal ponds, with a final

concentration of 14%. The latter process includes MEA extraction, compression, dehydration

and transportation, and produces a final concentration of almost 100%. Marine water was

provided as the water supply.

The microalgae production system is shown in 10.2. Algae was grown in raceway ponds,

and used solar drying to achieve a biomass that could be used for co‐firing.

Figure 10.2: Simplified flow diagram for microalgae production in Kadam’s study

Source: (Kadam, 2002)


Coordination Action


91

12.1.5 Results

Four impact assessment categories were evaluated: greenhouse warming potential,

depletion of natural resources, air acidification potential, and eutrophication potential. Algae

co‐firing was shown to have lower greenhouse potential than a coal fired power plant, as well

as a lower air acidification potential. The CO2 emitted was lower, as well as the SOx and NOx,

particulates, methane, and the fossil energy consumption. However, the depletion of natural

resources (more gas and oil is necessary) and eutrophication potential (due to the fertilizer)

was much higher for algae co‐firing. As such, any final decision on the usefulness of algae co‐

firing would have to find the correct balance between these factors (Kadam, 2001; 2002).

12.1.6 Discussion

This study does not focus on algae for biofuels, but does provide a useful look at possible

linkages that could be formed between algae production and industry – especially in the use

of using flue gas from power stations to grow algae.

Many of the individual assumptions in this study have been called into question (such as

productivity values) (Benemann, 2010b), while the use of solar drying remains a large

question, as it may not be possible (certainly not in all parts of the world) and experts agree

that it probably won’t be feasible in large scale processes (Sander & Murthy, 2010; Clarens,

2010b). Dewatering was not shown to be a large factor in the design, even though other

studies show that it is.

Also, the study discusses the eutrophication potential, but unlike agriculture, algae

cultivation is done under controlled circumstances which limit the possibility of a run‐off (Leu,

2010).

12.2 Lardon et al. 2009

Lardon et al. (2009) state that since this LCA is not based upon actual systems, it should be

used more to identify possible bottlenecks and:

“The key objective of this study is not to offer a LCA of the current microalgal biodiesel

technology, but to identify the obstacles and limitations which should receive specific research

efforts to make this process environmentally sustainable.”

No specific location was used as a basis for this study.


Coordination Action


92


The functional unit was the combustion of 1MJ of fuel in a diesel engine (assumed to act

the same way as other biofuels), and the boundaries included the extraction and production

of raw materials, facility construction (assuming a 30 year lifespan of buildings, 10 years for

electrical engines, etc), and biofuel elaboration and use in the engine.


Due to the lack of large scale commercial algae producers, the data was extrapolated from

laboratory scale experiments. The data was collected from industrial partners, inventory data

of similar processes, the EcoInvent database and literature reviews.

The required nitrogen inputs, as well as the heating value of the oil were calculated based

upon the algae composition, and standard practices from pilot scale systems were adapted as

necessary to fit the larger system (i.e. centrifugation works well on small scale but is too

expensive for larger scales).

12.2.4 Process

The hypothetical system is shown in 10.3, and consists of an open pond raceway system

covering 100ha, using Chlorella vulgaris. Two methods of operation were considered: normal

levels of Nitrogen, and low N fed to the system. Harvesting is done by use of continuous

circulation of the algae through a thickener, and then the flocculated stream is dewatered.

Two methods of oil extraction were evaluated, namely: advanced drying followed by hexane

extraction (as used for soybeans) (dry), or direct extraction from the wet algal paste (wet).


Coordination Action


93

Figure 10.3: Process chain overview from Lardon’s study

Source: (Lardon et al., 2009)

12.2.5 Results

The LCI was done without any allocation, but reflects the actual flows in the process chain,

as shown in Table 10.1. A cumulative energy analysis was performed to find the cumulative

energy demand (CED), which showed that only the wet extraction on low‐N grown algae has a

positive balance. Reducing the amount of nitrogen fed to the system improves the CED by 60%

(due to the lower fertilizer requirements), whereas wet extraction only increases the CED by

25% (since it needs larger production due to the lower extraction yield).


Coordination Action


94

Table 10.1: Most important material and energy flows generated by the production

of 1kg of biodiesel from Lardon et al’s study

Normal Low N

Dry Wet Dry Wet

Algae culture and harvesting

Algae (kg) 5.93 8.39 2.7 3.81

CO2 (kg) 10.4 14.8 5.32 7.52

Electricity (MJ) 7.5 10.6 4 5.7

CaNO3, as g N 273 386 29.4 41.6

Drying

Heat (MJ) 81.8 37.1

Electricity (MJ) 8.52 3.9

Oil extraction

Heat (MJ) 7.1 22.4 3.2 10.2

Electricity (MJ) 1.5 8.4 0.7 3.9

Hexane loss (g) 15.2 55 6.9 25

Oil transesterification

Methanol (g) 114 114 114 114

Heat (MJ) 0.9 0.9 0.9 0.9

Total Energy

Consumption (MJ) 106.4 41.4 48.9 19.8

Production (MJ) 103.8 146.8 61 86

Balance (MJ) ‐2.6 105 12 66


The impact analysis was done and focused on: abiotic depletion (AbD), potential

acidification (Ac), eutrophication (Eu), global warming potential (GWP100 (time horizon 100

years)), ozone layer depletion (Ozone), human (HumTox) and marine (Mar Tox) toxicity, land

competition (Land), ionizing radiations (Rad) and photochemical oxidation (Photo). The results

are shown in 10.4.

Algae was compared to other biofuels, and shown to be better for eutrophication, and land

use, but worse for ionizing radiation, photochemical oxidation and marine toxicity. The energy


Coordination Action


95

demand is also shown to be highly dependent on the process used, which should be kept in

mind for future designs (Lardon et al., 2009).

Figure 10.4: Comparison of impacts categories in Lardon et al ’s study


12.2.6 Discussion

This study clearly shows that drying is one of the most energy intensive steps. It

demonstrates what a difference in nutrient supply would mean to the final outcome, as well

as some of the different possibilities in downstream processing of the algal biomass. However,

experts have been critical about this study, since many of the values were based on pilot scale

systems, and would not be wholly applicable to industrial cultivation (Leu, 2010).

12.3 Clarens et.al. (2010)

In this study, the life cycle of the cultivation of algae was compared to corn, switch grass

and canola. The study was based in Virginia, Iowa and California in the US, each of which have

different levels of solar radiation and water availability.


The functional unit was 317 GJ, equivalent to the approximate per capita primary energy

consumption of one American.


Coordination Action


96

The system was limited to cultivation and harvesting only, with the final product being dry

biomass. It was assumed that the biomass was burned directly so the higher heating value

(HHV) was used to calculate the energy obtained. This was done because of the uncertainties

surrounding the choice of conversion process and the final product (electricity vs. liquid fuels).

The manufacture of the equipment used was considered negligible and not included.


Data was collected from previously published pilot‐scale demonstration projects, climactic

records, the EcoInvent database, and various other sources (as far as possible based in the

United States), and input using the Crystal Ball® predictive modelling suite. For example, the

dry biomass yields were calculated based upon radiation and meteorological data for the last

30 years, as well as empirical estimates for the radiation use efficiency (RUE).

12.3.4 Process

The chosen cultivation method for the algae was open raceways with paddle wheels for

mixing, with harvesting done by flocculation and centrifugation. Three different types of

wastewater effluent were considered as nutrient sources: activated sludge, source‐separated

urine, and biological nitrogen removal. The process, as well as those of the other feedstocks

considered, is shown in Figure 10.5.


Coordination Action


97

Figure 10.5: Schematic of system considered in Clarens study

Source: (Clarens et al., 2010)

12.3.5 Results

The output of the model was five impact categories: energy consumption (MJ), water use

(m3), greenhouse gas emissions (kg CO2 equivalent), land use (ha), and eutrophication (kg PO4

eq.). Algae cultivation was found to emit more GHG’s than it sequesters, requiring more fossil‐

based carbon to produce the same amount of bioenergy, while corn, canola and switch grass

had a net uptake of CO2. Water use for algae was also considerably higher than the others.

Energy production is positive for all of the studied biofuels in that they generate more energy

than they consume. Algae was however approximately 3.3–5 times more efficient in the land

use than the competitors. It is calculated that it would only require 13% of the USA’s land area

to meet the country’s total energy needs. The eutrophication (calculated on the basis of any


Coordination Action


98

potential leakages/spillages and any contributions from fertilizers) was also considerably lower

for algae. The values of a system based in Virginia, USA, are shown in Table 10.2.

Table 10.2: Life cycle burdens of Algae, Corn, Canola, and Switch grass in Virginia

Land (ha)

Energy (MJ)

x 104

GHG (kg CO2

equiv.) x 104

Water (m3)

x 104

Eutrophication

(kg PO4 equiv)

Algae 0.4 ± 0.05 30 ± 6.6 1.8 ± 0.58 12 ± 2.4 3.3 ± 0.86

Corn 1.3 ± 0.3 3.8 ± 0.35 ‐ 2.6 ± 0.09 0.82 ± 0.19 26 ± 5.4

Canola 2.0 ± 0.2 7.0 ± 0.83 ‐ 1.6 ± 0.10 1.0 ± 0.14 28 ± 5.8

Switch grass 1.7 ± 0.4 2.9 ± 0.27 ‐ 2.4 ± 0.18 0.57 ± 0.21 6.1 ± 1.7

Source: (Clarens et al., 2010)

Sunlight and water availability was not shown to have such a large impact on the final

outcome (California, as the sunniest was shown to have a slightly lower land requirement than

the other places, but a slightly higher energy requirement). The principle burden of algae

cultivation is the fertilizers (about 50% of energy use and GHG emissions).

The potential of co‐firing with coal power plants, as Kadam suggested, could provide large

reduction in the GHG, but would require large scale cultivation, which would increase the

fertilizer demand such that it would still be larger than that of corn, canola, or switch grass.

Using wastewater to supplant the need for chemical fertilizers (as well as water use)

substantially reduced the environmental impact of algae cultivation as well as the treatment of

the wastewater itself (meaning there is incentive for both algae producers and wastewater

treatment plants to work together) (Clarens et al., 2010).

12.3.6 Discussion

This is one of the more widely known LCA studies on algae, so there has been more

discussion surrounding it. Criticisms were made of the paper by Subhadra (2010), Benemann

(2010b) and others, including:

• The rationale of using Radiation Use Efficiency (RUE) instead of the absolute bio‐

productivity is not clear;

• The choice of functional unit (1MJ of algae biomass is not equivalent to 1MJ of switch

grass biomass as the former can be more easily converted to biofuels);

• The age of the data;


Coordination Action


99

• High fertilizer energy input as compared to other studies, as well as high cost

estimation of CO2 usage;

• Clarens et al. assumed that photobioreactors would not give a better land footprint,

but would give a higher life cycle burden – these assumptions were not fully justified

in his study;

• As with Kadam (2001), the eutrophication potential is limited in algae production as it

is a controlled, closed system;

• The manufacturing burdens should decrease in the future as electricity, the

manufacturing process, and the know how all becomes better and more

environmentally sustainable; and,

• The LCA performed does not take a holistic view of the system, but views each factor

as mutually exclusive (i.e. not taking into account that algae can produce co‐products,

etc.).

Subhadra (2010) concludes by stating:

“The investors are eagerly reviewing the feasibility status of various feedstocks, so studies

such as this are crucial. However, defining a comprehensive LCA boundary with an integrated

account of indirect GHG emission, energetic costs, and other sustainability indexes would yield

more meaningful comparisons”.

Clarens (2010a) responded to these criticisms by stating that although the actual system

boundaries used can be disputed, overall is should not make a great difference to the final

result. He justifies not including the co‐products in his LCA because of the varied nature and

amount of these co‐products, which would not fit in with the generic LCA performed. He

justifies the data by saying he updated it as necessary, as well as only focusing on existing,

rather than emerging technologies which could be more sustainable. He defends the choice of

open ponds over PBRs by claiming (as others have done) that PBRs are not yet commercially

viable. The eutrophication potential is calculated on the basis of the possibility of a leakage

from the closed system, not from continuous run‐off. Finally, he states that individual aspects

would not alter the overall outcome of the LCA, as shown by sensitivity analysis that he

performed in the LCA.

This study does not paint algae in the most positive light (as other biofuels are considered

more sustainable), but has received a lot of attention in the media, especially when taken out

of context. This could be damaging to the industry as a whole (Greenwell, 2010).


Coordination Action


100

12.4 Jorquera et.al. (2010)

The aim of this study was to compare the energy life‐cycle of the production of oil rich

microalgae, using the net energy ratio (NER) as a comparison factor, where NER is given as:

(Eqn 1.)

The study did not look at the final GHG emissions, only at the energy requirements. No

specific location was used for the study.


The functional unit was a production level of 100,000 kg of biomass (dry weight) produced

under various production methods. The assessment was performed using the GaBi® software,

which estimates the output of a particular process (in this case biomass produced and energy

generated) from the input of the energy costs associated with each process (i.e. the price of

raw materials, transportation, and equipment used).

Each systems unit (shown in Figure 10.6) was considered to be made up of the raw

materials required, the transport, and the manufacturing of the equipment. The inoculation

steps, oil extraction, and conversion to biofuels were not considered.

Figure 10.6: System Process in Jorquera’s study

Source: (Jorquera et al 2010). The red line indicates the system boundary


Coordination Action


101

12.4.3 Results

The results of the analysis are shown Table 10.3.

Table 10.3: Comparative analysis of three cultivation methods from Jorquera’s study

Variable Raceway Ponds Flat‐plate PBR Tubular PBR

Annual biomass production (kg/yr) 100,000 100,000 100,000

Volumetric productivity (kg / m3 day) 0.035 0.27 0.56

Space required (m2) 25,988.25 10,147.00 10,763.20

Reactor volume (m3) 7,827.79 1,014.71 489.24

Flow rate to maintain dilatation rate (m3 / day) 782.79 101.47 48.9

Relative oil content (%) 29.6 29.6 29.6

Net oil yield (m3 / yr) 32.9 32.9 32.9

Oil yield per area (m3 / ha year) 12.65 31.6 30.56

Energy consumption (W / m3) 3.72 53 2500

Energy consumption (W) 29,119.37 53,779.80 1,223,091.98

Total energy consumption (KWh / months) 8,735.81 16,133.94 366,927.6

Total energy consumption (GJ / year) 378.45 698.94 15,895.8

Total energy content in biomass (GJ / year) 3,155.30 3,155.30 3,155.30

NER for oil production 3.05 1.65 0.07

NER for biomass production 8.34 4.51 0.20

Source: (Jorquera et al., 2010)

Due to evaporation, the water consumption for raceway ponds is more than 16 times

higher than tubular PBRs, and 7 times higher than flat plate. The productivity and biomass

concentration is also significantly higher in the PBR systems, while the land requirements are

lower. However, the energy requirement, and final NER is quite a bit lower for raceway ponds,

as well as the construction and materials costs.

The results indicate that raceway and flat‐plate PBR systems had a NER > 1, and are

therefore economically viable. However, since the extraction and conversion was not included,

this could significantly add to the costs and energy requirements. It is clear from this study

that tubular PBR systems are not currently feasible at a large scale (Jorquera et al., 2010).

12.4.4 Discussion

This is a relatively unknown study, and quite limited in its scope. A major flaw of this study

is that it only focuses on the cultivation, and only looks at the energetics without due

consideration of the GHG emissions, economics, etc. As the other studies show, harvesting is a


Coordination Action


102

major component of the final energy balance, and the overall downstream processing will

have a large impact on the final energy consumption.

It shows the comparative land requirements to produce the same amount of biomass for

the three systems (showing how raceway ponds would need almost 2.5 times more land than

PBR systems). However, the data was based on pilot scale information, so does not take into

account the economies of scale.

12.5 Sander& Murthy (2010)

This was a well‐to‐pump study, done to determine the overall sustainability of algae biodiesel.

The objective of the study is given:

“Understanding the environmental burdens of the algal biodiesel production will allow

insight into inherent sustainability. Information from this LCA will be useful in identifying

energy and emission bottlenecks in the process. This information can be used to provide

impetus for further technological advancement of algal biodiesel and reduce overall energy

use and environmental impact of a future algal biodiesel process” (Sander & Murthy, 2010).

Since a lot of the data was based in the US, it is assumed that that is the location of the

study.


The functional unit was 1,000 MJ energy from algal biodiesel using existing technology

(which is equivalent to 24kg of algal biodiesel).

The system boundaries were determined by the RMEE protocol, discussed earlier, using a

cut off ratio of 5%, and shown in Figure 10.7.

This LCA deals with co‐product allocation by using the system expansion method. It

assumes that the carbohydrate and protein in the algal biomass will be used as feedstock for

ethanol conversion process, and by doing so, offset currently used feedstock (corn). It is

assumed to have the same ethanol yield as wheat straw due to the similar glucan content.


Data was gathered from the Greenhouse Gases, Regulated Emissions and Energy use in

Transportation (GREET) model, the US LCI database, as well as the 1998 Sheehan et al. report

and a 1988 Borowitzka & Borowitzka paper for information from the growth to separation

stage, as well as a few other specialized papers.


Coordination Action


103

12.5.4 Process

The process was based on the current state of the art technology upon which it is assumed

future systems will base themselves on. The process starts with inoculum grown in PBRs,

which is then transferred to an open pond raceway. Treated wastewater is the given medium,

presumed to provide all the necessary nutrients except for carbon, which is provided

separately (flue gas sparging is not considered in this study). Separation and dewatering then

occurs, with two methods considered: filtration and centrifugation. The final steps are hexane

extraction, with the resulting transesterification occurring at a different site and following the

same method as previously reported soybean conversion to biodiesel. Finally, the biodiesel is

transported to the pump, as described by the GREET model. The process diagram is given in

.

12.5.5 Results

The energy requirements and emissions for each step are given in Error! Reference source

not found.. It is clear that using a centrifuge requires more energy, while the most energy and

GHG emission intensive step is that of harvesting. Natural gas drying of the algal cake requires

69% of the total energy input – solar drying can significantly reduce this but is impractical at a

large scale and is not suitable for all climates. It is clear to see that the downstream processes

are important for the final sustainability of the system, and improvements must be made to

enhance the performance of algae as biofuels (Sander & Murthy, 2010).


Coordination Action


104

Figure 10.7: Process flow diagram from Sander & Murthy’s study

Source: (Sander & Murthy, 2010)


Coordination Action


105

Figure 10.8: Energy and Emissions associated with unit process in Sander &

Murthy’s study

Source: (Sander & Murthy, 2010)


Coordination Action


106

12.5.6 Discussion

One of the strengths of the Sander & Murthy (2010) study was a detailed overview of

where they obtained their data values from, as well as a systematic approach to defining the

system boundaries. They clearly indicated what values were used, and described the

limitations and changes they made to the data. However, as with the other studies discussed,

a major issue is the validity of the assumptions used (some of the papers are 22 years old, and

the data from the LCI databases are not specialized for algae cultivation).

12.6 Stephenson et.al. (2010)

This study is a well‐to‐pump analysis, with sensitivity analysis on various operating parameters

included. The basis of the study is in the UK, which has lower solar radiation than the other

papers considered.


The functional unit of this system was 1 ton of biodiesel, blended with conventional diesel

for a use in a compact sized car.

The boundaries include all the background systems, or homogenous markets, providing the

materials and energy to the main process. It was assumed that the land used would be

derelict, so there would be no system that the algae cultivation is replacing. The lifetime of the

equipment was assumed to be 20 years, and maintenance impacts were considered negligible.

By‐products of the system include algal residue, glycerol and potassium phosphate. The

algal residue is assumed to be anaerobically digested onsite to generate methane and satisfy

the heating requirements of the process, with excess sent to generate electricity. In this case,

direct substitution for the electricity was used. The glycerol can be used by the pharmaceutical

industry, using the market price as an allocation method. Another possibility is using it for the

generation of heat, where direct substitution could be used. The last by‐product is relatively

small in quantity, so the market price was used as an allocation method.


Coordination Action


107


Certain assumptions were made based on commercial and pilot scale systems currently in

use, as well as previous experience of the authors in growing algae. Other values were

obtained from literature.

12.6.4 Process

The basic system is shown in Figure 10.9. Two systems were considered, a raceway pond

and an air‐lift tubular PBR. For each system, there were two stages of production – the first

stage under a nutrient‐sufficient environment to achieve a dense culture, and a second stage

with no nitrogen supply to increase the triacylglycerides (TAG) content of the culture to

approximately 40%. Once cultivated, the algae is dewatered by flocculants (using Aluminium

Sulphate), with the spent medium returned to the cultivation. The biomass then undergoes

homogenization to break open the cells, and the lipids are extracted using hexane. This would

then be refined using the same method to recover rapeseed oil, before being transported to a

facility to convert it, via transesterification, to biodiesel, which is then transported for final use

in a car. It is assumed that the conversion to biodiesel occurs in a pre‐existing plant (previously

described in other papers) that can produce 250,000 tons of biodiesel per year. The aqueous

waste stream would be anaerobically digested to convert it into methane.

Various modifications to this system were considered. These include altering some of the

operating parameters, the TAG content, and changing the mode and distance of transport

required between process units.


Coordination Action


108

Figure 10.9: Process chain for production of 1 ton biodiesel in Stephenson et al’s

study

Source: (Stephenson et al., 2010)

12.6.5 Results

The results of the base case are shown in Figure 10.10. It is clear that tubular PBR requires

more energy, and as such has a higher Global Warming Potential (GWP). Compared to fossil‐

derived diesel, the energy requirements and GWP are 85 and 78% lower respectively for


Coordination Action


109

raceway ponds, but 362 and 273% higher respectively for tubular PBRs. It is also clear to see

that cultivation is the largest component of energy usage, with the electricity accounting for

85% of the energy use (including manufacture) for the air‐lift tubular PBRs. For raceway

systems, this was 74% of the energy use for that process unit, with the manufacture of the

PVC lining the biggest single contribution.

Figure 10.10: LCA results for base case production of biodiesel from Stephenson et

al’s study

Source: (Stephenson et al., 2010)

Water usage was estimated at 3.8 and 13.7m2/ton of biodiesel for raceways and PBR

systems. Raceway systems are assumed to need less water as the amount of rainfall in UK

exceeds the expected evaporation. In other countries, the expected water usage would

therefore be higher.

Modifications to the operating parameters showed where savings could be made as well:

i.e. lower velocities and recycling of the nutrients would lower the GWP burden, while use of

enzyme disruption would increase it. Some factors, such as the method of oil extraction, the

distribution of the biodiesel, and the final use of glycerol has little impact on the final results9

(Stephenson et al., 2010).

9


Coordination Action


110

12.6.6 Discussion

This system is localized for the UK, which gives a better view of the country impacts on the

algae growth (i.e. the water usage). This differs from other studies, which tended to focus

more on sunnier climates and areas of less rainfall. This was the only study to focus on some

of the possible operating parameters that could be improved upon (although there are many

more to be considered), which is useful for future considerations of the design of a facility.

This was also the only study which used primary data for the system, but limited the pond size

to 100m2, which limits the final outcome. However, it did mean that the assumptions were

based more firmly on what is possible rather than hypothetical, and many of the experts

consulted agree that this study uses more valid assumptions (Benemann, 2010b; Greenwell,

2010).

12.7 Campbell et.al. (2010)

This study looked at the environmental impacts of growing algae in ponds. The location of this

study was in Australia, which has a high solar incidence, but limited fresh water supply.


The functional unit was the “the combustion of enough fuel in an articulated truck (AT; the

most common form of freight transport in Australia) diesel engine to transport one tonne of

freight one kilometre, i.e. a tonne kilometre (t km or tkm). This has previously been calculated

to require the equivalent of 0.89 MJ of diesel fuel, which is 23.057 ml of ULS diesel” (Campbell

et al., 2010).

The system boundaries exclude the production facilities and construction, partly because

detailed information about all the subsystems is not yet readily available (and is also rarely

considered in analysis of fossil fuels). For the rest, the study looks to do a complete cradle to

grave analysis, with the waste algal biomass used to generate methane.


The data for the system was gathered from literature, and is mainly based on the 1996

Benemann and Oswald paper, as well as a paper in 1983 by Regan and Gartside (outlining the

design). Other literature, as well as the Australian LCI and the relevant Australian authorities

(such as the department of energy), were consulted for details and updating of the system. If


Coordination Action


111

data was not available, the closest information available from the EcoInvent database was

used. The system was run using SimaPro 7® software.

12.7.4 Process

The proposed system covers 400 ha of raceway ponds over 500 ha of (non‐arable, arid)

coastal land (the current usage of these lands is not mentioned or considered), with water

supplied directly from the ocean (after use and treatment, it is then returned to the ocean).

After cultivation, chemical flocculants are added to the system, followed by dissolved air

flotation to concentrate the sludge. It is then heated and centrifuged to extract the lipids and

concentrate the solution. The lipids are then transesterified in a process similar to that used

for canola (and other vegetable oils). For this reason, an algal species with similar lipid content

to canola was chosen (approximately 43% (Barthet, 2010)). The remaining algal mass is then

anaerobically digested to produce methane, which is then used to generate electricity.

Various scenarios were run using this model, including two different productivities (a base

case, and an optimistic case), and different methods of introducing the CO2. This includes

direct injection from an ammonia plant, flue gas from a power plant, or a liquefied form

bought commercially.

12.7.5 Results

The results for the two different productivities are given in Table 10.5. As a comparison, the

values for canola and ultra low sulphur (ULS) diesel, which is the standard of diesel used at the

time of the study, is included.

Table 10.5: GHG emissions (kg CO2 equivalent) for 1 ton km truck use

Impact category

Biodiesel, algal, 100% CO2 (ammonia plant)

Biodiesel, algal, 15% CO2 (flue gas) power station

Biodiesel, algal, 100% CO2 (truck delivered)

Biodiesel, canola

ULS diesel

15 g/m2/d

GHG gas (total fossil) ‐15.648 ‐10.524 18.164 35.856 81.239

GHG (total fossil upstream)

‐16.118 ‐10.994 17.695 35.386 19.213

GHG (total fossil tailpipe)

0.470 0.470 0.470 0.470 62.026

GHG (total upstream) ‐16.037 ‐10.913 17.719 35.465 19.241


Coordination Action


112

GHG (total tailpipe) 62.028 62.028 62.028 62.028 62.026

GHG ‐ CO2 (fossil upstream)

‐16.742 ‐11.477 16.594 33.437 18.040

GHG ‐ CO2 (fossil tailpipe)

0.001 0.001 0.001 0.001 61.557

GHG ‐ CH4 (total) 0.102 ‐0.033 0.364 0.984 1.156

CHG ‐ N20 (total) 0.990 0.984 1.138 1.431 0.486

GHG ‐ CO2 (total upstream)

‐16.689 ‐11.562 16.470 33.659 18.047

GHG ‐ CO2 (total tailpipe)

61.559 61.559 61.559 61.559 61.557

GHG ‐ other (total) 0.001 0.001 0.068 0.004 0.000

Total cost (Aus �) 4.3 3.9 4.8 4.2 3.8

30 g/m2/d

GHG gas (total fossil) ‐27.560 ‐23.019 8.298 35.856 81.239

GHG (total fossil upstream)

‐28.030 ‐23.489 7.828 35.386 19.213

GHG (total fossil tailpipe)

0.470 0.470 0.470 0.470 62.026

GHG (total upstream) ‐27.949 ‐23.408 7.852 35.465 19.241

GHG (total tailpipe) 62.028 62.028 62.028 62.028 62.026

GHG ‐ CO2 (fossil upstream)

‐28.563 ‐23.797 6.861 33.437 18.040

GHG ‐ CO2 (fossil tailpipe)

0.001 0.001 0.001 0.001 61.557

GHG ‐ CH4 (total) 0.030 ‐0.181 0.250 0.984 1.156

CHG ‐ N20 (total) 0.971 0.957 1.117 1.431 0.486

GHG ‐ CO2 (total upstream)

‐28.547 ‐23.987 6.657 33.659 18.047

GHG ‐ CO2 (total tailpipe)

61.559 61.559 61.559 61.559 61.557

GHG ‐ other (total) 0.001 0.001 0.068 0.004 0.000

Total cost (Aus �) 2.8 2.2 3.0 4.2 3.8

Source: Campbell et al 2010


Coordination Action


113

Note: Fossil emissions are from the combustion of fossil fuels and add extra GHG to the

atmosphere. Non‐fossil emissions are those that come from burning biomass, and are simply

releasing Carbon that had been previously fixed by the biomass – as such they add no new

Carbon to the atmosphere. Upstream emissions are those released during processing, while

tailpipe emissions come from combustion in the truck.

The negative emissions are because the biomass is anaerobically digested to produce

methane, then electricity, directly substituting electricity generation from other sources. The

results show that biodiesel from algae is better than canola and standard diesel, with savings

between 63.1 and 108.8 g / ton / km to ULS possible.

The study also looked at the economic considerations, although this is rife with uncertainty

(due to changing fuel prices, possible tax exemptions, changing landscape of the legislation,

etc.). The amortization rate was given at 15%, with the capital costs accounting for 43‐64% of

the annual operating costs. The conclusion is that under the right conditions, biodiesel from

algae is economically and sustainably feasible.

12.7.6 Discussion

This is the only other LCA study (other than Kadam) that uses marine water (fresh water is

scarce in Australia). However, a major disadvantage is that the environmental considerations

of desalinating and then treating this water are not considered (Greenwell, 2010). The study is

specific to Australia, as it has the distinct advantages of receiving more sun than other

countries, while also having more land available for such facilities without a big impact on

biodiversity.

Although the economic considerations are very basic calculations, with the assumptions

subject to change, it does give a favourable view of the economic feasibility of algae under the

right conditions, and is the only study to look at this aspect as well as the technical side.


Coordination Action


114

13. Annex 3: Expert Stakeholders participating in this study

Name Association Position and experience Type of response

John Benemann

Benemann Associates, USA

Researcher – formerly with the NREL, many years of experience in algae

Personal communication and completed questionnaire

Tomas Branyik1

Institute of Chemical Technology, Czech Republic

Researcher – current projects include bioethanol from algae

Completed life cycle inventory spreadsheet

Andres Clarens University of Virginia, USA

Researcher – lead author of a recent LCA study Completed questionnaire

Chris Greenwell2

Durham University, UK

Researcher – previous work includes feasibility studies on algae, industry consultant, and various (EU) projects

Completed questionnaire

Diana Fonseca1 Necton – AlgaFuel, Portugal

Chemical Engineer ‐ has design and production experience with algae

Completed questionnaire

Stefan Leu Ben Gurion University, Israel

Researcher – experience with biotechnology and sustainability

Personal communication and discussion

Vitor Verdelho1

Necton – AlgaFuel, Portugal

R&D Manager at Necton with experience in microalgae biotechnology


Pierre‐Antoine Vernon1,2

European Biodiesel Board (EBB)

Project Manager – experience in lobbying for legislation with biofuels


1 Member of AquaFUELs 2 Member of EABA


Coordination Action


115

14. Annex 4: Questionnaire

Current LCA studies

We have identified only 7 major LCA studies that provide an overview of the energy balance

for algal biofuels (listed above)

• Please list any important studies that you think we have missed.

• Are you familiar with any of the studies, Do you have any concerns about their

credibility – and if so what are they?

• Do you think these LCA provide a good reflection of the merits of algal biofuels

o If yes – how could they be improved

o If no – why not?

Click here to enter text.

What, in your view, are the main strengths and weaknesses of LCA (for algae)?


Algae for biofuels

Which production pathway (cultivation through to final conversion to biofuel) do you believe

is most likely to be commercial, and why? Should all the various production pathways be

considered in LCAs?


Current LCA studies identify water consumption for algae cultivation to be an important issue,

and recommend using wastewater. Do you think this will help determine where algae biofuels

can be grown and which water source do you think has the greatest potential (i.e. fresh,

marine or waste)?


• What other factors will determine where the production plants are located (what land

constraints are there in terms of proximity to a carbon, water, solar radiation source,

type of land)



Coordination Action


116

What, in your view, are the most important technical challenges that have to be overcome for

successful commercial cultivation of algae for biofuels?


LCA on algae

Is there currently a lack of good quality data and does this affect the overall reliability of a LCA

(current studies use theoretical data based on lab/pilot scale)?

• Are there currently too many unknowns to make LCA results useful for guiding

decisions?


The inputs for current LCAs are estimated based on the composition of the algae – whereas

inefficiencies are more likely in scaled up production processes. Does this mean that current

studies too optimistic?


Co‐products and services can improve the calculated sustainability of algae – how are these

best incorporated in LCA studies (i.e. if algae are concurrently used for wastewater treatment

and biofuel production)?

• Which are the most likely co‐products to be used and for what?

• Do you have a preferred method of allocation, and if so, why?


“CO2 use by algal cultures is not CO2 sequestration – that comes from algal biofuels replacing

fossil fuels”. There is some debate about the carbon credits that can be assigned to algae

cultivation, with policy makers taking the position that carbon captured by algae growth is

later released anyway, so does not count as carbon sequestration. Under what circumstances

do you think CO2 credits for sequestration can be attributed to algal biofuels?



Coordination Action


117

Different LCAs adopt different functional units, such as impacts per 1000MJ, the amount of

biofuel to transport 1 tonne by 1 km, algae grown in 1 ha, etc. Given the uncertainty of the

production pathway, which is the most logical functional unit that can be adopted for future

LCAs in order to maintain consistency between studies, and how important is it to have a

consistent functional unit between studies?


Different LCAs use different system boundaries ‐ – is there any merit to adopting a standard

practice (i.e. Relative Mass Energy and Economics method proposed by Raynolds) for all future

LCAs – or where do you see the logical system boundary should be?


Are there any further issues you think should be considered when using LCA to guide policy

decisions in the future?



Coordination Action


118

15. Annex 5: Assumptions of Normalized Modelling

Primary Energe Usage Assumptions1.Algae Cultivation and Harvesting

Electricity used in Cultivation System Electricity consumption from the cultivation system operation is based on original LCA study

Nitrogen fertilizer

Nitrogen fertilizer demand is based on original LCA study; Energy Content of the Nitrogen fertilizer is based on Fehrenbach (2008)"GHG Accounting Methodology and Default Data according to the Biomass Sustainability Ordinance (BSO)"

Tractor Fuel usage Diesel tractor used in harvesting algae (Based on Campbell et al 2009)

System Construction Tubular manufacture and PVC lining (Based on Stephenson et al 2010)

2. Biomass Drying and Dewatering

Heat

Studies which assumed certain heat input in the Biomass drying and dewatering stage still follow their original assumptions; Studies which exlude the extraction process are normalised by Lardon et al's assumption

Electricity

Studies which assumed certain electricity input in the Biomass drying and dewatering stage still follow their original assumptions; Studies which exlude the extraction process are normalised by Lardon et al's assumption

3.Lipid Extraction Electricity (Homogenization for Cell disruption) Only applied on the Stephenson et al(2010)'s study

Electricity( Extraction)

Studies which assumed certain electricity input in the extraction stage still follow their original assumptions; Studies which exlude the extraction process are normalised by Steohenson et al's assumption

Heat

Studies which assumed certain heat input in the extraction stage still follow their original assumptions; Studies which exlude the extraction process are normalised by Steohenson et al's assumption

Totals1.Algae Cultivation and Harvesting Sum up all the energy input in this stage2.Biomass Drying and Dewatering Sum up all the energy input in this stage3.Lipid Extraction Sum up all the energy input in this stageBiomass Yield Based on original LCA studyBiorefinery Based on original LCA studyVolumetric Fuel yield Based on original LCA studyEnergetic fuel yield Based on original LCA studyTotal Heat Input Sum up all the Heat InputsTotal Electricity Input Sum up all the Electricity Inputs Total Input energy Sum up all the Energy Inputs

Total Energy Output (on Mass Basis) Energy content in the algal oil plus Energy output from Gas boiler. (on the mass basis)

Total Energy Output (on Energy Basis) Energy content in the algal oil plus Energy output from Gas boiler. (on the energetic basis)

Energy content in Biomass( Heating value of Biomass)

Studies which had assumed Heating value of the biomass are still follow their original assumptions; Studies which didn't mention the value, the energy content of the biomass is calculated by Illman et al(2000) 's results

Subtotal Energy Output in Lipid(Heating value of algal oil)

Studies which had assumed Heating value of the lipid are still follow their original assumptions; Studies which didn't mention the value, the energy content of the lipid is calculated by Illman et al(2000) 's results

Subtotal Energy Output in Residual Energy content in the Algal biomass minus energy content from the algal lipid

Energy Content in Methane(Anaerobic Digestion)(60%)

Methane yield is based on Sialve et al (2009)

Generated heat from Boiler(Assuming 75% efficiency)

Efficiency of the Gas boiler is based on Stephenson et al 2010

Heat Generation From CHP(41%) Electricity Generation From CHP(34%)

Reported HV of Algal Lipid

Studies which had assumed Heating value of the lipid are still follow their original assumptions; Studies which didn't mention the value, the energy content of the lipid are calculated by Illman et al(2000) 's results

Reported HV of Biomass

Studies which had assumed Heating value of the biomass are still follow their original assumptions; Studies which didn't mention the value, the energy content of the biomass are calculated by Illman et al(2000) 's results

Net Energy Ratio (NER) for Lipid production All the Primary Energy Input minus Energy content of coproducts, then divided by Energy content of lipid fraction

Net Energy Ratio (NER) for Biomass production All the Primary Energy Input divided by Energy content of dry biomass

Only applied on Stephenson et al 2010: System Efficiency is based on John Macadam (2010)


Coordination Action


119

16. Annex 6: Example of data normalization

Algae Cultivation and Harvesting

Original Unit Original Value

Normalized Unit Normalized Value

Electricity Consumption (MJ/kg Biodiesel) 7.50 (MJ/MJ Dry Algal Biomass)

0.0521

Land Use (m2) 1000000.00

(m2) 1000000.00

Transportation of inputs to farm

(km) 100.00 (km) 100.00

Nitrogen fertilizer (g/kg Biodiesel) 273.00 (MJ/MJ Dry Algal Biomass)

0.093

Phosphorus fertilizer (g/kg Dry Biomass) 9.90 (g/kg Dry Biomass) 9.90 CO2 Source (kg/kg Algal Biomass) 1.80 (kg/kg Algal Biomass) 1.80 Tractor Fuel usage (MJ/kg Dry Algal

Biomass) NR (MJ/MJ Dry Algal

Biomass) NR

System Construction (MJ/kg Biodiesel ) NR (MJ/MJ Dry Algal Biomass)

NR

1: Original data in Lardon et al 2010: lipid ratio: 17.5%; extraction efficiency: 70%. For every kg biodiesel

produced , 8.163kg biomass is required, data for electricity consumption was allocated to biodiesel production

which is 7.5MJ/kg Biodiesel, after normalization as all the electricity consumption allocated to biomass

production, the result will be 0.9188 MJ/kg biomass, the heating value of the biomass in Lardon et al’s study is

17.5 MJ/kg biomass for normal Nitrogen cultivation case, so we can get 0.052MJ/MJ biomass after data

normalization.( all the data with yellow colored are had been normalized)

079_D 3.3-3.5 Life-Cycle Assessment and Environmental Assessment

Documents

Transcript of 079_D 3.3-3.5 Life-Cycle Assessment and Environmental Assessment