BIOTECHNOLOGY LETTERS

Volume 18 No. 9 (September 1996) pp.1013-1018

Received as revised 8 July

ETHANOL PRODUCTION FROM CORN COB PRETREATED BY THE

AMMONIA STEEPING PROCESS USING GENETICALLY
ENGINEERED YEAST

N. J. Cao .1, M. S. Krishnan 1'2, J. X. Du 1, C. S. Gong 1,

N. W. Y. Ho ~, Z. D. Chen ~and G. T. Tsao ~'2
aLaboratory of Renewable Resources Engineering and 2School of Chemical Engineering
Purdue University, West Lafayette, IN 47907.

SUMMARY
A new and effective pretreatment process for biomass conversion involves the
steeping of biomass in 2.9 M Nt-hOH. This resulted in the removing about 80-90% of
the lignin along with almost all the acetate from cellulosic residues. Based on dry
cellulose from corn cob, a high glucose yield of 92% was obtained after enzymatic
saccharification of cellulose fraction. By using a genetically engineered, xylose-
fermenting Saccharomyces 1400(pLNH33) in the batch fermentation of a glucose-xylose
mixture from corn cob, an ethanol concentration of 47 g/L was obtained within 36 h with
84% yield. In addition, an ethanol concentration of 45 g/L was obtained within 48 h
with 86% yield using simultaneous saccharification-fermentation process.

INTRODUCTION
Lignocellulosic biomass is a complex mixture of cellulose, hemicellulose, lignin
along with extractives. Biomass pretreatment is therefore a necessary process operation
in order to achieve high ethanol yields. Grinding and milling are the primary physical
pretreatment operations that are too expensive. Chemical pretreatment can be
accomplished by using strong acids or bases. However, this process option is also
expensive and compounded with the necessity of chemical recovery. Steam explosion of
biomass is another pretreatment method that is effective but involves considerable
thermal energy in the form of steam. Moreover, the high temperatures involved also
cause degradation of the xylose that is formed by the hydrolysis of hemicellulose

1013
(McMillan, 1994). The low temperature, ammonia fiber explosion (AFEX) process
overcomes this problem by carrying out the explosion process at low temperature. The
efficiency of this pretreatment method is higher at higher pressures and under this
conditions part of the pentose fraction is lost with the lignin stream (Dale & Moreira,
1982).
In this paper we report a new pretreatment method that involves steeping of the
lignocellulosic biomass (using corn cob as a model feedstock) in dilute NI-LOH at
ambient temperature to remove lignin, acetate and extractives. This is followed by a
dilute acid treatment that readily hydrolyzes the hemiceUulose fraction to simple sugars,
primarily xylose. The residual cellulosic fraction of biomass can then be enzymatically
hydrolyzed to glucose. Hence, by using this pretreatment method in biomass processing
the lignin, hemicellulose and cellulose can be easily separated and processed separately.
Both the cellulose and hemicellulose hydrolyzates produced by this process can be
fermented to ethanol by suitable yeast. The solid cellulosic residue can also be subjected
to the simultaneous saccharification and fermentation (SSF) process for biomass
conversion to ethanol. It is known that the cellulase enzyme can be adsorbed by lignin,
thus decreasing the cellulase activity (Bernardez et al., 1993). By using the
NH3/HC1/SSF process, the lignin is separated from the biomass in the initial steeping
step. This enables a more effective use of the cellulase enzyme during the SSF process.
As the cost of the cellulase enzyme contributes significandy to the biomass conversion
process (Wyman, 1994), this processing option could make the biomass to ethanol
conversion more economically feasible.

EXPERIMENTAL METHODS
Materials:
Corn cob milled to 3 mm, consisting of 41% cellulose, 36% xylan, 7% lignin, and 3.2%
acetate value, was purchased from Andersons Inc., Maumee, Ohio. The cellulase was
provided by Iogen Co., Canada and the specific activity of the enzyme as determined by
Iogen Co. was 170 IFPU/ml. Genetically engineered Saccharomyces 1400(pLNH33)
described by Ho, et al., (1993) and Saccharomyces1400 were used for this work.
Methods:
Ammonia steeping. Corn cob (20 g) was mixed with 100 ml 2.9 M NI-I4OH in a 250 ml
Erlenmeyer flask and was incubated in a shaker for 24 h at 26°C. The mixture was
filtered to separate the corn cob. Corn cob was then washed twice by deionized water to
remove residual ammonia from the surface of corn cob. The delignified corn cob
preparation was then obtained by vacuum evaporation to remove residual ammonia. The

1014
lignin can be precipitated out of solution upon ammonia removal and collected by
filtration. Ammonia was collected by a cooling trap for reuse.
Dilute acid hydrolysis. Delignified corn cob was treated with 0.3 M HC1 solution at
100-108°C for 1 h. Acidic hemicellulose hydrolyzate obtained was neutralized by NaOH
followed by desalting using IRA-94 anion-exchange resin. The pretreated cellulosic
residue was then washed with deionized water to remove residual acid.
Enzymatic hydrolysis of cellulosic residual. To the cellulosic residues obtained from
20 g of corn cob, 50 ml water and 1.0 ml cellulase enzyme (equivalent to 8.5 IFPU/g
corn cob) were added in a 250 ml flask. The hydrolysis was carried out at 50°C for 48 h.
Ethanol fermentation of hydrolyzates. The genetically engineered yeast
1400(pLNH33) was grown in YEPX medium (1% yeast extract, 2% Bacto-peptone, and
2% xylose). Fermentation of hemicellulose hydrolyzate or mixture of cellulose and
hemicellulose hydrolyzates was performed by mixing 50 ml hydrolyzate containing 2%
Bacto-peptone and 1% yeast extract with 0.5 ml concentrated yeast culture in 125 ml
flask at 30°C.
Simultaneous saccharification and fermentation (SSF) of cellulosic residue.
Pretreated corn cob was added to a 250 ml Erlenmeyer flask with 50 ml yeast extract-
peptone medium. It was autoclaved for 15 min at 121°C. The total volume of water in
the flask was 80 ml, after taking into account the water content of the pretreated corn
cob. The initial pH was adjusted to 5.5 by phosphate buffer. Following this, 0.5 ml
concentrated yeast culture and cellulase (8.5 IFPU/g corn cob) were added to initiate
SSF of the cellulosic residues at 35°C in a shaker at 250 rpm. Samples (0.2 ml) were
taken at 12 h interval over 108 h of incubation.
Analysis. Glucose, xylose, ethanol, acetate, glycerol and xylitol were analyzed by HPLC
(column: 300x7.8 mm HPX-87H, Bio-Rad, Richmond, California). Lignin was
determined as Klason lignin by weight method. The residual ammonium ion in solution
was determined by 05800-05 Solution Analyzer equipped with an ammonium electrode
(Cole-Parmer Instrument Co., Niles, Illinois).

RESULTS AND DISCUSSION

In this pretreatment process, the lignocellulosic biomass was steeped in a dilute
ammonia solution at ambient temperature to extract and remove the lignin. This was
followed by hydrolysis using a dilute acid at 100-108°C to hydrolyze the hemicellulose,
and increase the surface area of cellulose available for the enzymatic hydrolysis in the
simultaneous saccharification and fermentation (SSF) process. Hence this was referred
to as the NH3/HCI/SSF process. The ammonia steeping also caused the cellulosic
fraction in the material to swell, thereby enhancing the effectiveness of the acid
hydrolysis step. This process differed from the ammonia fiber explosion (AFEX) process
both in concept and technique. Unlike the AFEX process, all the hemicellulose was kept
intact and was not subjected to degradation during lignin removal. The sequential

1015
separation of the major components (i.e., lignin, hemicellulose and cellulose) of
lignocellulosic biomass was the primary objective in this process.
Depending on the material under consideration, about 80-90% of the lignin and
almost all the acetate and alkali-soluble extractives were removed from the raw material
by steeping it in 2.9 M ammonia solution at 26°C. The removal of the above
components from lignocellulosic feedstocks was a critical step in the ethanol production
process, as these components had been identified as the inhibitors during the
fermentation of sugars (especially xylose) by yeast (Du Preez, 1994). Based on our
studies with the above process, the solubility of lignin in dilute ammonia solution was
much higher than reported elsewhere. However, different steeping conditions may be
required for materials having different lignin content in order to carry out an effective
pretreatment. Following the ammonia steeping, the ammonia can be removed under
vacuum at below 60°C with a recovery yield of 98%. A high quality lignin can be
isolated from the steeped extract. This lignin by-product, unlike those from acid
hydrolysis or pulping processes, was pure. This allows the use of the isolated lignin in
synthesis of polymers and chemicals.
The results of enzymatic hydrolysis of corn cob treated by different techniques is
shown in Figure 1. Results indicated the combination of ammonia steeping followed by
dilute acid hydrolysis gave the highest glucose yield of 92% based on dry cellulose.
Following the dilute acid hydrolysis, a cellulose rich residue was available. Since all the
lignin, acetate, alkali extractives and hemicellulose have been removed by prior process
steps, a lower enzyme dosage was required for effective hydrolysis of the cellulose to
glucose. Moreover, the glucose-rich solution obtained did not require further treatment
and can be directly used as a substrate for ethanol production.
Following ammonia steeping, corn cob hemicellulosic fraction can be hydrolyzed
readily by dilute acid and separates from the cellulosic fraction. The hemicellulose
hydrolyzate (rich in xylose) obtained has no acetate and alkali extractives present that
can be used as a substrate for xylose fermenting yeast. Before the fermentation, the salt
was removed from the hemicellulose hydrolyzate by treating it with a weak-based anion
exchange resin. As shown in Figure 2, Saccharomyces 1400 (pLNH33) can readily
ferment hemicellulose hydrolyzate containing 5 lg xylose/L of to produce 20g ethanol/L
within 36 hours at a yield of 80% (based on the theoretical yield of 0.51 g ethanol per g

1016
 60
100t 2.9 M Ammonia~.3 M HCI, 0.3 M HCI

2.9 M Ammonia Control

4 o" "'- e I
4t- -4t- 50

80
~,40

60
30

40
20

20
lO

0
0 10 20 30 40 50 0 10 20 30 40 50
Time (h) Time (h)
Figure 1. EnzymaticHydrolysisTime Course of Corn Cob Figure 2. Fermentation of Hemicellulose Hydrolyzate
Pretreated by Different Process. Yield:92% for Ammonia/ from Corn Cob Using Yeast 1400(pLNH33)
HCi Process; 64% for Dilute HC1;57% for Ammonia
60 60
1. Ammonia/HCl2. HCI only 3. Ammoniaonly Control I

50 ,-. 50

,-,40 6~.se
O

g 30
xyn..m.~l
Gl~:bcer°l
20

10 10

0 it
0
o i0 20 30 40 50 60 0 20 40 60 80 100 120
Time (h) Time(h)
Figure 3. Fermentation of Hemicellulose-Celluiose Figure 4. SSF Ethanol Production from Corn Cob after
Hydrolyzate Mixture from Corn Cob Using Yeast Different Treatment. Yield: 1--86%; 2=72.3%; 3=68%
1400(pLNH33)

1017
xylose). In the experiment where the hydrolyzate was not treated with the ion exchange
resin, the fermentation time was much longer (72 h) and the ethanol yield was lower
(~50%). The same yeast can also ferment both glucose and xylose simultaneously. This
is demonstrated in Figure 3, where a sugar mixture consisting of 52.8 g glucose/L and
55.7 g xylose/L was fermented to 46.9 g ethanol/L within 36 h, giving a high yield of
84% as based on the theoretical value.
The SSF process has been identified to be an economically viable option for
conversion of biomass to ethanol (Philippidis and Smith, 1995). The SSF process
improves the hydrolysis rates and ethanol yields in comparison to processes involving
separate hydrolysis and fermentation. In order to demonstrate the effectiveness of our
pretreatment process, SSF experiments were performed using yeast strain
Saccharomyces 1400 and corn cob samples, pretreated by using different techniques as
shown in Figure 4. The combination of ammonia steeping followed by dilute acid
hydrolysis again gave the best results. An ethanol concentration of 45 g/L was obtained
within 48 hours with a yield of 86% based on dry cellulose from corn cob. Application
of this pretreatment process also allowed the SSF process to be performed with a low
cellulase loading (8.5 IFPU/g corn cob) indicating that the pretreatment process was
generally applicable, irrespective of the fermenting microorganism.

ACKNOWLEDGMENTS
This study was funded through The Consortium for Plant Biotechnology
Research, Inc. by DOE cooperative agreement no. DE-FCO5-92OR22072 and no. DE-
FCO5-92OR22072A and Great Lakes Regional Biomass Energy Program.

REFERENCES
Bernardez, T. D., Lyford, K., Hogsett, D. A. and Lynd, L. R. (1993). Biotechnol
Bioeng., 42, 899-908.
Dale, B. E. and Moreira, M. J. (1982). Biotechnol. Bioeng. Symp., 12, 13-20.
Du Preez, J. C. (1994). Enzyme Microb. Technol., 16, 944-56.
Ho, N. W. Y., Chen, Z. D. and Brainard, A. (1993). Genetically Engineered Yeasts
Capable of Effective Fermentation of Xylose to Ethanol. In: Proceedings of Tenth
International Symposium on Alcohol Fuels. Colorado Spring, Co, pp.738-745.
McMillan, J. D. (1994). Pretreatment of Lignocellulosic Biomass. In: Enzymic
Conversion of Biomass for Fuels Production. M. E. Himmel, J. O. Baker and R. P.
Overend, eds. pp. 292-324, ACS, Washington, DC.
Philippidis, G. P. and Smith, T. K. (1995). Appl. Biochem. Biotechnol., 51/52, 117-24.
Wyman, C. E. (1994). Bioresource Technol., 50, 3-16.

1018