全 文 ：第 31 卷第 6 期
2011 年 12 月
林 产 化 学 与 工 业
Chemistry and Industry of Forest Products
Vol． 31 No． 6
Dec． 2011
A Two-step Statistical Optimization of Cultivation
Conditions for Producing Lipid by
Scenedesmus obliquus
收稿日期:2011 － 03 － 02
基金项目:林业公益性行业专项(201004001)
作者简介:杨 静(1985 －) ，女，河北保定人，硕士，主要从事生物质能源方面的研究;E-mail:yj3219@ 163． com
* 通讯作者:蒋剑春，研究员，博士，博士生导师，研究领域为林产化学加工和生物质能源开发技术;E-mail:bio-energy@ 163． com。
YANG Jing
YANG Jing，JIANG Jian-chun* ，ZHANG Ning，WEI Min
(Institute of Chemical Industry of Forest Products，CAF;National Engineering Lab． for Biomass
Chemical Utilization;Key and Open Lab． on Forest Chemical Engineering，SFA;Key Lab． of
Biomass Energy and Material，Jiangsu Province，Nanjing 210042，China)
Abstract:A fast and easy two-step statistical optimization of cultivation condition for producing lipid by
Scenedesmus obliquus was described． At first，a fractional factorial design (FFD)was used to screen sig-
ni-ficant factors from six factors (illumination，temperature，pH value，inoculation，glucose concentration and nitrogen source
concentration)． Then an orthogonal experiment was executed to further determine optimal conditions． According to the results，
nitrogen source concentration and temperature were selected as significant factors． The optimal conditions occurred at nitrogen
source concentration 6． 0 mmol /L，temperature 25 ℃ and illumination 80 μmol /(m2·s) ，with the highest lipid productivity of
75． 46 mg /(L·d)．
Key words:Scenedesmus obliquus;fractional factorial design;orthogonal design;lipid productivity
CLC numbers:TQ35;TQ517 Document code:A Article ID:0253 － 2417(2011)06 － 0007 － 04
两步实验设计优化栅藻产油条件的工艺研究
杨 静，蒋剑春，张 宁，卫 民
(中国林业科学研究院林产化学工业研究所;生物质化学利用国家工程实验室;国家林业局林产化学工程
重点开放性实验室;江苏省生物质能源与材料重点实验室，江苏 南京 210042)
摘 要:提出了一种简单、快速优化栅藻(Scenedesmus obliquus)产油条件的两步实验设计方法。首先，利用部分析因设计
方法从 6 个因素(光照强度，温度，pH值，接种量，葡萄糖浓度，氮源浓度)中筛选出显著因素。再利用正交试验进一步确
定出最优条件。结果表明:氮源浓度和温度是影响栅藻产油的显著因素。适合培养栅藻产油的最优条件为:氮源
6． 0 mmol /L，温度 25 ℃ 和光照 80 μmol /(m2· s) ，最高油脂产率达到 75． 46 mg /(L·d)。
关键词:栅藻;部分析因设计;正交试验;油脂产率
Continued use of petroleum-sourced fuels results in energy crisis and environmental degradation． Renew-
able，clean biomass energy has become a research hotspot，in order to achieve economic and environmental
sustainability． Biomass refers to plant，animal or microbial life on the material synthesized by the general
term． Biomass instead of fossil resources in the development of chemical industry is the only way of sustainable
human development［1］． Efficient development and utilization of biomass energy will play a very active role in
addressing energy and environmental issues［2］． Biodiesel is a kind of long-chain fatty acids，produced by
transesterifying biomass oil to achieve a lipid of viscosity close to that of petrodiesel［3］． It is recognized as a
8 林 产 化 学 与 工 业 第 31 卷
potential renewable and carbon neutral alternative to petroleum fuels［4］． Microalgae have gained much atten-
tion as sources of biodiesel，due to its high photosynthetic and breeding rates，strong environmental adaptabili-
ty，as well as easy integration with other engineering［5］．
Not all species of algae are suitable for biodiesel production，it requires high lipid content as well as fast
growth rate［6］． Scenedesmus sp． is a kind of green algae． The average oil content in Scenedesmus sp． varies be-
tween 12 % and 22 %［7－8］ of dry weight(few strains can reach over 70 %［9］) ，which is lower than some high-
lipid algae，such as Chlorella sp．(28 %－32 %)［4］． However Scenedesmus was reported as an adequate fatty
acid profile for biodiesel production，due to its low linolenic and polyunsaturated contents［7］． Besides，
Scenedesmus has long been utilized as bioremediature agents in the removal of organic or inorganic nutrients
from polluted water，which means that this sort of wastewater can be used as culture medium for Scenedesmus
and thus can overcome the high cost of biodiesel production［10］． So lipid production from Scenedesmus is worth
paying attention to．
The aim of this work is to optimize cultural conditions for producing lipid by Scenedesmus obliquus through
a two-step statistical method． A fractional factorial design (FFD)was chosen to screen the significant factors，
and then，the optimum values were determined using orthogonal experiment．
1 Material and Methods
1． 1 Strain and growth medium
Scenedesmus obliquus (obtained from Institute of Hydrobiology，Chinese Academy of Sciences)was
initially cultured in a medium containing(per liter) :0． 25 g NaNO3(3 mmol /L) ，0． 075 g K2HPO4，0． 075 g
MgSO4·7H2O，0． 025 g CaCl2·2H2O，0． 175 g KH2PO4，0． 025 g NaCl，0． 005 g FeCl3·6H2O，1 mL
EDTA－Fe，1 mL trace metal solution (2． 86 g H3BO3，1． 86 g MnCl2·4H2O，0． 22 g ZnSO4·7H2O，0． 39 g
Na2MoO4·2H2O，0． 08 g CuSO4·5H2O，0． 05 g Co(NO3)2·6H2O per liter) ，40 mL soil extraction，with
10 g /L glucose，at 23 ℃，under continuous illumination with a cool fluorescent lamp at 50 μmol /(m2·s)．
The cells were cultivated in 250 mL flasks with 100 mL medium．
1． 2 Lipids extraction
Biomass was lyophilized to constant weight after centrifugation． According to the method described by Wu
et al［11］，lipids were extracted by solvent-extraction method with petroleum ether and ethyl ether (volume frac-
tion 2 ∶ 1)． The lipid productivity WLP(mg /(L·d) )was calculated as follows:
WLP = Y·W/t
Where:W—the accumulated dry cell weight from inoculation to harvest，mg /L;Y— the lipid content，%;t—
the overall cultural time，d．
1． 3 Fractional factorial design
A matrix 2 ×(6 － 2)for FFD experiments is presented in Table 1． Two levels (－ 1 and + 1)for each
variable:illumination，temperature，pH value，inoculation，glucose concentration and nitrogen source con-
centration (X1 － X6)． Lipid productivity was chosen as response． All analysis was performed using the STA-
TISTICA 8． 0 Software．
1． 4 Orthogonal experiments
Based on the results of FFD，an orthogonal L9(3
4)test design was designed to further optimize the condi-
tions for lipid productivity．
2 Results and Discussions
Using the FFD，16 runs were designed as the 2 ×(6 － 2)in duplicates (Table 1)． Table 2 shows the es-
第 6 期 杨 静，等:两步实验设计优化栅藻产油条件的工艺研究 9
timated effect of variables on the responses． Nitrogen source concentration and temperature with small P-levels
(＜ 0． 05)were selected as significant factors． Nitrogen source concentration has been reported to contribute to
high lipid content，but the biomass growth rate is slowed down because of the inhibition of cell division resul-
ting from the environmental stresses［12］． The order of the other four factors was:illumination，glucose concen-
tration，pH value and inoculation． Besides，the effects of pH value and inoculation were negative，which indi-
cates that lower pH value and inoculation are benefit for lipid production，while other four factors exhibit posi-
tive effects． Therefore，nitrogen source concentration，temperature as well as the third influence—illumination
(for requirement of orthogonal experiment)were selected for further optimization． Inoculation and pH value
were fixed at 10 % (volume fraction)and 5 respectively． Glucose concentration was fixed at 10 g /L from the
economic view of point．
Table 1 Detailed design parameters and results of the 2 ×(6 －2)FFD experiments on the lipid
productivity by Scenedesmus obliquus
No．
X1
illumination /
(μmol·m－2·s－1)
X2
temperature /℃
X3
pH value
X4
glucose
concentration /
(g·L－1)
X5
inoculation
(volume fraction)/%
X6
nitrogen source
concentration /
(mmol·L－1)
WLP
lipid productivity
/(mg·L－1·d－1)
1 － 1(40) － 1(25) － 1(5) － 1(10) － 1(10) － 1(1． 5) 27． 81
2 + 1(80) － 1(25) － 1(5) － 1(10) + 1(20) － 1(1． 5) 23． 45
3 － 1(40) + 1(28) － 1(5) － 1(10) + 1(20) + 1(3． 0) 38． 59
4 + 1(80) + 1(28) － 1(5) － 1(10) － 1(10) + 1(3． 0) 44． 34
5 － 1(40) － 1(25) + 1(8) － 1(10) + 1(20) + 1(3． 0) 19． 21
6 + 1(80) － 1(25) + 1(8) － 1(10) － 1(10) + 1(3． 0) 28． 40
7 － 1(40) + 1(28) + 1(8) － 1(10) － 1(10) － 1(1． 5) 18． 33
8 + 1(80) + 1(28) + 1(8) － 1(10) + 1(20) － 1(1． 5) 29． 45
9 － 1(40) － 1(25) － 1(5) + 1(20) + 1(20) + 1(3． 0) 23． 97
10 + 1(80) － 1(25) － 1(5) + 1(20) + 1(20) + 1(3． 0) 45． 35
11 － 1(40) + 1(28) － 1(5) + 1(20) + 1(20) － 1(1． 5) 19． 68
12 + 1(80) + 1(28) － 1(5) + 1(20) － 1(10) － 1(1． 5) 40． 97
13 － 1(40) － 1(25) + 1(8) + 1(20) + 1(20) － 1(1． 5) 17． 39
14 + 1(80) － 1(25) + 1(8) + 1(20) － 1(10) － 1(1． 5) 13． 15
15 － 1(40) + 1(28) + 1(8) + 1(20) － 1(10) + 1(3． 0) 50． 13
16 + 1(80) + 1(28) + 1(8) + 1(20) + 1(20) + 1(3． 0) 53． 82
Table 2 Results analysis of the 2 ×(6 －2)FFD experiments on the lipid productivity by Scenedesmus obliquus
factors effect coef． t-value P-value
illumination /(μmol·m－2·s－1) 7． 977 3． 988 1． 820 0． 102
temperature /℃ 12． 073 6． 036 2． 755 0． 022
pH value － 4． 285 － 2． 143 － 0． 977 0． 354
glucose concentration /(g·L－1) 4． 360 2． 180 0． 994 0． 346
inoculation(volume fraction)/% － 0． 020 － 0． 010 － 0． 004 0． 996
nitrogen source concentration /(mmol·L－1) 14． 198 7． 098 3． 240 0． 010
Based on the results of FFD，an orthogonal design L9(3
4)was conducted to optimize the conditions for
lipid productivity． The factors correspond to nitrogen source concentration (A) ，temperature (B)and illumi-
nation(C)． As shown in Table 3，the orders of effects on lipid productivity were:nitrogen source concentra-
tion ＞ temperature ＞ illumination． The optimal conditions for lipid productivity were A3B2C3，i． e． nitrogen
source concentration，temperature and illumination were 6． 0 mmol /L，25 ℃ and 80 μmol /(m2·s) ，respec-
10 林 产 化 学 与 工 业 第 31 卷
tively． Under this condition，the highest lipid productivity was 75． 46 mg /(L·d)．
Table 3 Orthogonal experiment design used to optimize culture conditions for lipid productivity by S． obliquus
No．
A
nitrogen source
concentration /(mmol·L－1)
B
temperature /℃
C
illumination /(μmol·m－2·s－1)
WLP
lipid productivity /
(mg·L－1·d－1)
1 1． 5 23 40 35． 66
2 1． 5 25 60 35． 67
3 1． 5 28 80 48． 76
4 3． 0 23 60 47． 56
5 3． 0 25 80 58． 79
6 3． 0 28 40 45． 79
7 6． 0 23 80 52． 75
8 6． 0 25 40 68． 46
9 6． 0 28 60 65． 16
k1 40． 03 45． 32 49． 97
k2 50． 71 54． 30 49． 46
k3 62． 12 52． 23 53． 43
R 22． 09 8． 98 3． 97
3 Conclusions
A two-step statistical method (fractional factorial design and orthogonal experiments)was used to opti-
mize the conditions for producing lipid by Scenedesmus obliquus． Nitrogen source concentration and temperature
were selected as significant factors． The highest lipid productivity of 75． 46 mg /(L·d)was obtained under ni-
trogen source concentration 6． 0 mmol /L，temperature 25 ℃ and illumination 80 μmol /(m2·s)of the initial
conditions．
References:
［1］宋湛谦．生物质资源与林产化工［J］．林产化学与工业，2005，25(增刊) :10－15．
［2］蒋剑春．生物质能源应用研究现状与发展前景［J］．林产化学与工业，2002，22(2) :75－81．
［3］DEMIRBAS A． Importance of biodiesel as transportation fuel［J］． Energy Policy，2007，35 (9) :4661－4670．
［4］CHISTI Y． Biodiesel from microalgae［J］． Biotechnol Adv，2007，25:294－306．
［5］梅洪，张成，殷大聪，等．利用微藻生产可再生能源研究概况［J］．武汉植物学研究，2008，26(6) :650－660．
［6］HU Q，SOMMERFELD M，JARVIS E，et al ． Microalgal triacylglycerols as feedstocks for biofuel production． Perspectives and advances［J］．
Plant J，2008，54(4) :621－639．
［7］GOUVEIA L，OLIVEIRA A C． Microalgae as a raw material for biofuels production［J］． J Ind Microbiol Biotechnol，2009，36 (2) :269－274．
［8］SPOLAORE P，CLAIRE JC，DURAN E，et al． Commol /Lercial applications of microalgae［J］． J Biosci Bioeng，2006，101:87－96．
［9］MATSUNAGA T，MATSUNOTO M，MAEDA Y，et al． Characterization of marine microalga，Scenedesmus sp． strain JPCC GA0024 toward biofuel
production［J］． Biotechnol Lett，2009，31:1367－1372．
［10］KIM M K，PARK J W，PARK C S，et al ． Enhanced production of Scenedesmus spp．(green microalgae)using a new medium containing fermen-
ted swine wastewater［J］． Bioresour Technol，2007，98:2220－2228．
［11］吴庆，蔡昭玲，丛威，等．从微藻中提取多不饱和脂肪酸［J］．北京化工大学学报，2004，31(4) :5－8．
［12］LV J M，CHENG L H，XU X H ． Enhanced lipid production of Chlorella vulgaris by adjustment of cultivation conditions［J］． Bioresour Techn-
ol，2010，101:6797－6804．