1

An Industrial Method of Biofuel Production from Chitin: Moving Towards Profitability

Abstract

Chitin, the long chain biopolymer of N-acetyl-D-glucosamine (GlcNAc), is the most

abundant biopolymer found in nature, after cellulose. However, chitin has largely been ignored

as a potential carbon source in biofuel production. We developed a complete input-to-output

industrial process of converting chitin to useful fuels and solvents, namely acetone, butanol and

ethanol (ABE). We developed procedures for the three steps of biofuel production: pretreatment,

hydrolysis and fermentation. For chitin pretreatment, we used a chitin extraction technique by

Takiguchi et. al consisting of dilute HCl demineralization, dilute NaOH deproteinization and

absolute acetone depigmentation. This improved pretreatment increased average chitin yield

from 3.2% in preliminary trials to 21.2%. For chitin hydrolysis, we found that 12M HCl and

12M H2SO4 hydrolyzed up to 94% of chitin to glucosamine (GlcN) compared with much lower

conversion rates with 6M and 3M concentrations of both acids. For fermentation of GlcN, we

developed an ABE process using bacteria of the Clostridium genus with empirically optimized

bacteria species and strain, sterilization procedure and fermentation length. Butanol production

reached an average of 8.578 g/L when using Clostridium acetobutylicum ATCC 55025, which is

competitive with glucose fermentation. All results were compounded into the process modeling

program SuperPro Designer for economic analysis. Using this model, we found that a large-scale

chitinous biofuel plant using our method would be near profitable with a gross profit margin of -

6.82%.

Keywords: biofuel, chitin, Clostridium acetobutylicum, pretreatment, hydrolysis, fermentation

2

Table of Figures

Table 1. Pretreatment percent yield chitin, protein and CaCO3 from crab shell………………7

Figure 1. Percent hydrolysis of chitin to GlcN by various methods…………………………...8

Figure 2. Butanol production by C. acetobutylicum ATCC 824, C. acetobutylicum ATCC 55025,

C. beijerinckii BA101, C. beijerinckii SV6 and C.tyrobutyricum……………………………...9

Figure 3. pH values of batch fermentations of C. acetobutylicum ATCC 55025 and 824……..9

Figure 4. Concentration of inhibitors FMA, LA, HMF in GlcN medium after sterilization by

various methods………………………………………………………………………………..10

Table 2. Comparison of average product concentration of C. acetobutylicum 55025 in various

mediums………………………………………………………………………………………..11

Figure 5. Screenshots of model of industrial chitinous biofuel production plant……………...12

3

Introduction

Research on biofuels has increased greatly in recent years as the price of non-renewable

fuels increases [6]. Currently, highly invested in areas include starch and cellulose-derived

bioalcohols. Starch-derived ethanol is the most common biofuel in the world and its production

provides 10% of American transportation energy consumption [6]. However, a major problem

with starch-based fuel production is starches are derived from food crops like corn. This places

pressure on already strained food sources, water sources and arable land. As a result, cellulose-

derived ethanol has received increased attention, so much so that several plants are already open

in the United States and are predicted to produce over 20 million gallons of ethanol in 2014 [5].

In many ways, chitin, a polysaccharide found in the exoskeleton of all organisms with an

outer skeleton and in the cell walls of microorganisms such as fungi, yeast and algae, could also

be an excellent candidate for conversion into biofuel. However, little research has been

conducted on the feasibility of chitin as a source of biofuel [10]. Chitin’s biofuel potential lies in

the fact that it is very similar to cellulose, which is already a promising source of biofuel.

Chemically, the two compounds are very alike, as chitin can be described as cellulose with one

hydroxyl group on each monomer replaced with an acetyl amine group. In terms of abundance,

chitin ranks just behind cellulose, which is the world’s most abundant biopolymer [9]. In

addition, millions of tons of chitin, in the form of arthropod shells, are discarded as waste in the

fishing industry, leading to a major environmental hazard [1]. Countless other sources of chitin

including fungi, which make up 25% of Earth’s biomass, also go unutilized. A method of

converting this chitin into biofuel would produce huge amounts of energy from an otherwise

untapped source, as well as solving an environmental problem by utilizing fishing waste.

4

The goal of this research project is to determine the feasibility of chitinous biofuels and

work towards economic viability. We hypothesized that several species and strains of solvent-

producing bacteria could produce valuable solvents from glucosamine (GlcN), the de-N-

acetylated monomer of chitin, including C. acetobutylicum, C. beijerinckii and C. tyrobutyricum,

which are known to produce acetone, butanol and ethanol from glucose. Combined with a

publically available chitin pretreatment method and an acid hydrolysis method, our fermentation

method would be the first start-to-finish model of chitinous biofuel production.

Materials and Methods

1. Pretreatment- Extraction of chitin from crab shell, a natural source of chitin (Method of

Takiguchi et. al) [7]

Preparation of Crab Shell. Crushed crab shell was obtained online through Amazon.com. The

shell was sonicated in distilled water-filled flasks for 30 minutes. The water was drained and the

process repeated 6 times. The shell was dried in the incubator (45°C) for 3 days.

Demineralization of Crab Shell. Crab shell was grinded using a coffee grinder. The shell was

slowly added to 1M HCl at 15:1 (v/w) solvent to solid ratio. This was done at ambient

temperature with constant stirring. Demineralization lasted 1h. The liquid was filtered using 25

µm filter paper. Then, the residue was washed several times with distilled water until pH=4,

dried in incubator for 3 days, then massed.

Deproteinization of Crab Shell. The dried, demineralized crab shell was massed along with

residue from filter paper. Then, dilute NaOH (3.5%) at a solvent to solid ratio of 10:1 (v/w) was

added to crab shell. The mixture was constantly stirred and maintained at a temperature of 65°C

5

for 2 hours. The residue was washed several times with distilled water until pH=9 and dried in

incubator for 3 days, then massed.

Removal of Carotenoids and Other Pigments. The dried, demineralized and deproteinized crab

shell was massed. Then, absolute acetone was added at a solvent to solid ratio of 10:1 (v/w)

under a fume hood and constantly stirred for 15 min at ambient temperature. The residue was

washed 10 times with distilled water and dried in incubator. After drying, the resulting chitin was

massed and stored in plastic test tubes.

2. Hydrolysis of Chitin

Chitin was added to HCl or H2SO4 solutions at 4:1 (v/w) in 100 mL test tubes. The test tubes

were sealed and placed in the hot water bath for 2h at 85°C. 1 mL supernatant sample was taken

for HPLC analysis. A vacuum pump was used for the evaporative removal of HCl.

3. Fermentation of Glucosamine (GlcN)

Making Nitrogen and Mineral Sources. The following solutions were created by adding the listed

chemicals to distilled water in their respective concentrations in 125 mL serum bottles. Oxygen

was purged from the bottles using N2 gas for 5 min. The serum bottles were sealed with rubber

stoppers and aluminum caps and sterilized by autoclaving at 120°C, 15 psig, for 30 min.

I. Nitrogen Source (100mL at 10x conc.) - K2HPO4 0.5 g/L, KH2PO4 0.5 g/L, NH4C2H3O2

2.2 g/L

II. Mineral Source (100mL at 200x conc.)- MgSO4•7H2O 0.1 g/L, FeSO4•7H2O 0.015 g/L,

CaCl2•2H2O 0.015 g/L, MnSO4•H2O 0.01 g/L, CoCl2•6H2O 0.02 g/L, ZnSO4•7H2O 0.002

g/L

6

Making Fermentation Mediums. The following solutions were created in 125mL serum bottles

with distilled water. Oxygen was purged from the bottles using N2 gas for 5 min and the serum

bottles were sealed with rubber stoppers and aluminum caps. After the listed method of

sterilization, 5 mL Nitrogen Source and .25 mL Mineral Source were added to each solution by

needle and syringe to create fermentation medium. All concentrations listed are for the final

50mL.

I. Cell Growth Medium (50 mL)- Sterilization by autoclave (120°C, 15 psig, 30

min)

a. 45mL Cell Growth Medium base solution- Glucose 30 g/L, (NH4)2SO4 2 g/L,

K2HPO4 1 g/L, KH2PO4 0.5 g/L, Tryptone 2 g/L, Yeast extract 1 g/L

II. P2 Medium (50 mL)- Sterilization by autoclave (120°C, 15 psig, 30 min)

a. 45mL P2 base solution- Glucose 60 g/L, Yeast extract 1 g/L

III. P2 w/GlcN Medium (50 mL)- Sterilization by autoclave (120°C, 15 psig, 30 min)

or filter (200 nm pore size)

a. 45mL P2 w/GlcN base solution- GlcN 60 g/L, Yeast extract 1 g/L

Fermentation. Once the solutions were created, the seed culture was incubated. Various bacterial

seed cultures were obtained from the Ohio State University Koffolt Labs bioengineering freezer

under Professor ST Yang. The cultures were thawed at room temperature and 1 mL of each were

injected into the Cell Growth Medium through the rubber stopper of the serum bottle using a

needle and syringe. This was done while working over a Bunsen burner flame. The seed cultures

were incubated at 37° C for 16 hours.

After 16 hours, 2.5 mL seed culture was taken from the serum bottle using a needle and

syringe and then injected into both the P2 Medium and P2 w/GlcN Medium. The medium was

7

then incubated at 40° C for 144 h, with 1 mL samples taken by needle and syringe every 24

hours. The pH of these samples was measured using a pH meter and adjusted with NaOH (3M),

which was injected into the serum bottles to maintain optimal pH.

4. Analysis

Gas Chromatography (GC). The solvent and organic acid concentrations of the samples were

analyzed using gas chromatography. The samples were centrifuged (5 min, 13000 rpm), and .05

mL of supernatant liquid was added to .95mL GC buffer. Then, the samples were mixed using a

vortex mixer. Finally, the samples were run through the gas chromatographer (Agilent 5977A

Series GC/MSD) for results.

High Performance Liquid Chromatography (HPLC). The sugar and inhibitor concentrations of

the samples were analyzed using the HPLC (Shimadzu RID-10A). The samples were centrifuged

(5 min, 13000 rpm), and added to GC buffer. Then, the samples were mixed using a vortex

mixer. Finally, the samples were run through the HPLC for results.

Results

Table 1. Pretreatment percent yield chitin, protein and CaCO3 from crab shell

Xavier method Takiguchi method trial 1 Takiguchi method trial 2

% CaCO3 0 68.4 64.7

% Protein 0 10.6 14

% Chitin 3.2 21 21.3

Pretreatment of a natural source of chitin, crab shell, was attempted with two different

methods found in literature. At first, chitin extraction was attempted using 6M HCl only, as

described by Xavier [11]. However, chitin yield from crab shell was low, at 3.2% (Table 1).

8

Then, pretreatment was attempted by a modified method of Takiguchi et al [7].

Demineralization, deproteinization and depigmentation involved the same methodology as

Takiguchi but sodium hypochlorite bleaching, a step used to improve the aesthetic quality of

chitin, was not conducted to prevent the creation of compounds toxic to bacteria. We yielded an

average of 21.2% chitin from crab shell in two trials, which validates the findings of Takiguchi et

al (Table 1) [7].

Hydrolysis of chitin using strong acid involves hydrolysis of the chitin to N-acetyl-D-

glucosamine and then de-N-acetylation to glucosamine [3]. GlcN concentration was measured

using HPLC to determine the percent conversion of chitin to GlcN. Hydrolysis was attempted

with HCl and H2SO4 at 12M, 6M and 3M. 12M HCl hydrolyzed the highest percentage of chitin

to GlcN, at 93.95% (Figure 1). 12M H2SO4 was close behind, hydrolyzing 90.23% chitin to GlcN

(Figure 1). 6M and 3M concentrations were far less effective for both acid types (Figure 1).

Three trials were conducted for each hydrolysis method.

0

10

20

30

40

50

60

70

80

90

100

12M HCl 6M HCl 3M HCl 12M H2SO4 6M H2SO4 3M H2SO4

Aver

age

% C

hit

in H

ydro

lysi

s to

Glc

N

Figure 1. Percent hydrolysis of chitin to GlcN by various methods. N=5

trials per method

9

Figure 2. Butanol production by C. acetobutylicum ATCC 824, C. acetobutylicum ATCC

55025, C. beijerinckii BA101, C. beijerinckii SV6 and C.tyrobutyricum. N= 5 trials. P2

signifies glucose fermentation, GlcN signifies glucosamine fermentation

Figure 3. pH values of batch fermentations of C. acetobutylicum ATCC 55025 and 824. N=5

trials. “P2” signifies glucose fermentation, “GlcN” signifies glucosamine fermentation, “filter”

signifies filtering with 200 nm filter was used as sterilization method.

We conducted preliminary fermentations using bacteria of the Clostridium genus,

specifically C. acetobutylicum ATCC 824, C. acetobutylicum ATCC 55025, C. beijerinckii

BA101, C. beijerinckii SV6 and C.tyrobutyricum. These bacteria are known to produce the

valuable solvents acetone, butanol and ethanol as well as the organic acids butyric acid and

0

1

2

3

4

5

6

7

8

9

P2 55025 GlcN 824 GlcN 55025 GlcN BA101 GlcN SV6 GlcN C.tyro

Buta

no

l p

rod

uct

ion (

g/L

)

Bacteria type

4.5

5

5.5

6

6.5

0 24 48 72 96 120 144 168

pH

Fermentation duration (hr.)

P2 824 P2 55025 GlcN 55025 GlcN Filter 55025

10

acetic acid from glucose and we predicted them to metabolize GlcN as well because of their

ability to metabolize a wide range of other sugars [14]. We focused on the bacteria’s production

of butanol since butanol is the most valuable biofuel of the three solvents and since the bacteria

produce more butanol than any other product.

C. acetobutylicum strains were found to have the highest butanol production from GlcN,

with the 55025 strain producing more butanol from GlcN than from P2, a commercial glucose

fermentation medium for the production of solvents using the Clostridium genus (Figure 2).

Taking into account that C. acetobutylicum 55025 produced more butanol from glucose than all

other Clostridium species and strains (not shown in Figure 2), preliminary GlcN trials using C.

acetobutylicum 55025 were very successful. But although C. acetobutylicum 55025 produces

more butanol from GlcN than from P2 on average, the average GlcN fermentation length is 48

hours longer than glucose fermentation, as the pH of glucose fermentations level off at 96 hours

while the pH of GlcN fermentations level off at 144 hours (Figure 3).

0

1

2

3

4

5

6

7

8

CaCO3 before

autoclave

CaCO3 before,

+NaOH

CaCO3 after autoclave filtered GlcN

Aver

age

Inhib

itor

conc.

(g/L

)

Figure 4. Concentration of inhibitors FMA, LA and HMF in GlcN

medium after sterilization by various methods. n=5 trials, "filtered" =

sterilization by 200 nm filter

FMA LA HMF

11

A major issue of GlcN fermentation was discovered when we found GlcN medium

degrades severely when autoclaved at 120° C for sterilization. The medium visibly developed a

black color and pressure buildup in the serum bottles. Since the degradation point of GlcN in

water is known to be much higher, at 150° C, we suspected a reaction between medium contents.

Neutralizing the acidic fermentation medium using NaOH before autoclaving did not lessen

degradation. However, when GlcN and CaCO3 were autoclaved separately instead of in the same

medium, GlcN degradation was visibly less severe. The GlcN medium had a light brown color

with no gas buildup, characteristics similar to P2 medium. So, it can be concluded that GlcN and

CaCO3 react at temperatures below normal GlcN’s degradation temperature, a new finding.

In addition, we measured the concentrations of three known fermentation inhibitors in the

fermentation medium using HPLC, alpha-monofluoromethylagatine (FMA), lactic acid (LA) and

5-hydroxymethyl furfural (HMF) [2]. We found much higher concentrations of FMA, LA and

HMF when CaCO3 was mixed with GlcN and then autoclaved than when CaCO3 and GlcN were

kept separate when autoclaving (Figure 4). This adds to the evidence that GlcN and CaCO3 react

at high temperatures.

Table 2. Comparison of average product concentration of C. acetobutylicum 55025 in

various mediums. N=6 trials, P2 signifies glucose fermentation

GlcN, filter

GlcN, +CaCO3

before autoclave

GlcN, +CaCO3

after autoclave P2

Acetone (g/L) 1.325 0.4845 1.197 2.008

Butanol (g/L) 8.578 1.587 5.766 7.950

Ethanol (g/L) 0.7122 0.4286 0.9562 0.6649

Butyrate (g/L) 1.507 1.843 1.757 2.222

Acetate (g/L) 3.542 2.677 3.522 3.734

Total Products

(g/L) 15.66 7.020 13.20 16.58

12

Six sets of fermentations were attempted on the original sterilization method medium, the

separated GlcN and CaCO3 medium and also a different “filtering” sterilization method (200 nm

filter, no autoclave) in order to determine the optimal sterilization procedure. We used C.

acetobutylicum 55025 since it had metabolized GlcN best and had produced the most solvents

from both GlcN and glucose. We found that the filtering sterilization method allowed for the

highest average concentration of useful products compared to the other two sterilization methods

(Table 2). In addition, we found that fermentations from filtered GlcN produce 7.90% more

butanol and 7.07% more ethanol than P2 glucose medium, which is a very optimistic result as C.

acetobutylicum 55025 was genetically engineered for glucose fermentations (Table 2) [14].

Figure 5. Screenshots of model of industrial chitinous biofuel production plant. Created

with SuperPro Designer®

13

We created a complete, start-to-finish model of a large scale chitinous biofuel plant using

the process modeling program SuperPro Designer® by Intelligen (Figure 5). The computer

model employs optimized pretreatment, hydrolysis and fermentation methods developed

empirically and applies them to a full-scale application, along with a common distillation

process. The model considers current-day prices for inputs and outputs of the process in order to

provide detailed reports on economic, materials and streams and environmental simulation data.

Table 3. Executive economic summary of chitinous biofuel production from fishing waste. Calculated by SuperPro Designer®

Total Capital Investment 29,072,000.00 $

Operating Cost 111,057,000.00 $/yr

Total Revenues 103,964,000.00 $/yr

Cost Basis Annual Rate 11,152,872.00 gal /yr

Gross Margin - 6.82 %

When scaled to an industrial size, the process developed in this study is near economic

profitability, with a gross margin of -6.82% (Table 3). From an annual input of 70,000 tons of

fishing waste, or roughly the waste output of an average-sized fishing port, the model plant

produces 11,152,872 gallons of acetone, butanol and ethanol a year (Table 3) [8]. This model is

the first chitinous biofuel production model ever engineered, and as a result holds valuable

lessons for the future of chitinous biofuel.

Discussion

Through this pioneering research, we have shown that chitin is a valuable potential

source of biofuel. After two years of work optimizing a novel start-to-finish process of chitinous

14

biofuel production, we have achieved near affordability according to our computer model of an

industrial plant. Perhaps our greatest achievements were in the glucosamine fermentation step,

where we were the first to identify certain members of the Clostridium genus to be metabolizers

of glucosamine, the first to optimize a fermentation procedure using glucosamine instead of

glucose and the first to document a reaction between glucosamine and CaCO3 at high

temperatures. In fact our optimization of glucosamine fermentation is so successful that C.

acetobutylicum produces more solvent from glucosamine than from a standard-procedure

glucose fermentation. Still, there are many improvements to be made to our method.

Our current method produces .261 g/g grams product produced per gram sugar consumed

whereas current industrial corn-based yeast-produced ethanol produces .48g/g grams ethanol

produced per gram sugar consumed [12]. Since butanol and acetone have higher energy densities

than ethanol, the efficiency discrepancy is not as large as may seem. Still, our method is clearly

not as efficient at converting sugar to fuel as current bioethanol production. In addition, we

empirically found that glucosamine fermentations take 48 h, or 50%, longer until completion

than glucose fermentations. This is another disadvantage of glucosamine fermentation as the rate

of production is lower than glucose fermentation as well.

Another issue with our method is the use of strong acids in the chitin hydrolysis step.

According to the computer model, nearly 50% of annual operating costs arise from the high cost

of HCl used in hydrolysis. In addition, strong acids such as HCl pose serious environmental

problems in terms of disposal. However, when we input the reaction rate of chitinases from the

bacterium Serratia marcescens at the same price as current cellulases, the process was profitable

according to our computer model. These chitinases show incredible promise in their efficiency

and high rate of catalysis, which are superior to the cellulases used in cellulosic ethanol

15

production [4,9]. Fortunately, there is an increasing amount of research on these chitinases to

decrease their price which could be transferred to our method [9].

In general, the great potential of chitinous biofuel production we have peered into should

be enough to sustain further research in the area. Although there are currently uses for chitin and

its derivatives, annual worldwide consumption does not exceed 15 thousand tons [3]. Annual

crustacean fishing waste alone amounts to 3 million tons of unused chitin [8]. Other sources of

chitin, including insects, plankton and yeasts produce a staggering 1010- 1011 tons a year [13]. So,

crustacean shell waste alone could potentially be converted to around 98 million gallons of

acetone, butanol and ethanol a year, a useful application for a waste source that is currently

dumped in landfills and oceans and creating an environmental problem [4]. Other easily

accessible sources of chitin like mayfly and locust swarms and krill are even more abundant and

thus could produce even more fuel.

Conclusion

We developed a three-part process of chitinous biofuel production. We used a publicly

available pretreatment process, a concentrated strong acid hydrolysis method and a C.

acetobutylicum fermentation to develop the first complete input-to-output chitinous biofuel

production method. Using data empirically obtained from this method, we created a computer

model to model a large-scale biofuel production plant using chitin as the carbon source. Our

computer model indicated that our method was nearly profitable. Further research on chitinases

for chitin hydrolysis and improving fermentation yield is needed to increase profitability of our

novel method.

16

References

1. Archer, M. (2013, January). Crustacea Processing Waste Management. Retrieved from

http://www.seafish.org/media/publications/fs53_6_10_crustacea_processing_waste_mana

gement.pdf

2. Ezeji, T., Qureshi, N., & Blascheck, H. P. (2006, July). Butanol production from

agricultural residues: Impact of degradation products on Clostridium beijerinckii growth

and butanol fermentation. Retrieved from University of Illinois website:

http://naldc.nal.usda.gov/download/2805/PDF

3. Einbu, A. (2007, April). Characterisation of Chitin and a Study of its Acid-Catalysed

Hydrolysis, Norwegian University of Science and Technology, 37(1): 12–87

4. Leffler, M. (1997, March/April). Treasure from trash: Is there profit in crab waste?

Maryland Marine Notes. Retrieved from ftp://ftp.mdsg.umd.edu/MDSG/

MarNotes/MN15_2.PDF

5. (2014, March) POET-DSM Commercial Cellulosic Bio-ethanol Plant. POET-DSM,

2014. Retrieved from http://poetdsm.com/pr/construction-remains-on-schedule-for-

project-liberty

6. Schnepf, R., & Yacobucci, B. D. (2013, March). Renewable Fuel Standard (RFS):

Overview and Issues. Retrieved from http://fas.org/sgp/crs/misc/R40155.pdf

7. Takiguchi, Y. et. al. (1991). Physical Properties of Chitinous Materials Chitin. In:

Advances in Chitin Science, Proceedings from the Asia-Pacific. R.H. Chen and H.C.

Chen (Eds), Chitosan-Jikken Manual, 3(1): 1-7.

8. (2012). The State of World Fisheries and Aquaculture. Retrieved from:

http://www.fao.org/docrep/016/i2727e/i2727e01.pdf

9. Vaaje-Kolstad, Gustav. (2013, March). The Chitinolytic Machinery of

Serratia marcescens – a Model System for Enzymatic Degradation of Recalcitrant

Polysaccharides. Wiley Online Library. John Wiley & Sons, Inc., 7 Mar. 2013. Retrieved

from http://onlinelibrary.wiley.com/doi/10.1111/febs.12181/full.

17

10. Wendland, Jürgen. N-Acetylglucosamine Utilization by Saccharomyces Cerevisiae Based

on Expression of Candida Albicans NAG Genes. Applied and Environmental

Microbiology (2009): 42(1): 75-84.

11. Xavier, Martin. "STANDARDIZATION OF OPTIMUM CONDITIONS FOR THE

PRODUCTION OF GLUCOSAMINE HYDROCHLORIDE FROM CHITIN." Indian

Council of Agricultural Research (2006)

12. Arshad, M. et. al. (2014, March). Kinetics and thermodynamics of ethanol production by

Saccharomyces cerevisiae MLD10 using molasses. Appl Biochem Biotechnol. 2014

Mar;172(5):2455-64.

13. Arbia, W. et. al. (2012, April). Chitin Extraction from Crustacean Shells Using

Biological Methods – A Review. Food Technol. Biotechnol. 51 (1) 12–25 (2013).

14. Qureshi, N., Blascheck HP. (2001) ABE production from corn: a recent economic

evaluation. Journal of Industrial Microbiology & Biotechnology (2001) 27, 292–297.