本研究的目的是评估12个不同的酿酒酵母菌株接种在连续乙醇发酵过程中酵母絮凝特性的持续能力。该系统是包括两个串联的无细胞循环反应器。进料底物是含有葡萄糖的合成培养基。评估的参数是进料底物的总还原糖，第一和第二反应器出口的酒精以及酵母的种类和数量。该系统达到83.53%的理论产率产量与最大总还原糖转化率达到92.68%。在这个系统的转化率低于预期。在这个过程两个反应器中酵母占主导地位,出乎意料的是酿酒CP6应变,无法形成球状尽管其絮凝增长。

 

关键词: 絮凝; 酵母; 生物燃料; 乙醇; 发酵

 

1 介绍

    使用生物燃料来代替石油是确保一个可持续未来最可行的方法。20世纪70年代的石油危机让巴西开发国家酒精计划，旨在取代汽油的替代燃料。在Proalcool的介绍下,研究人员鼓励发展项目，旨在减少这种生物燃料的生产成本。这种生物燃料的追求成本降低导致了发展的替代流程。在这些过程中使用的固定塔反应器中体现了酵母菌株的絮状特征。这种类型的过程,通过适当操作,可达到一个维持传统工艺获取同一收益率的高水平生产率。最近,更多的注意力已集中到固定化生物的活动,因为这种生物反应器的性能在发挥重要作用的工艺过程已经被承认(Gikas和利文斯顿,1997)。

    众多研究表明,乙醇的生产利用固定化酵母。（Joekes 等人研究(1998)）固定化酿酒酵母株到温石棉(Mg4Si4O10(OH)3)最后的收率为26%，高于自由细胞。（Wendhausen 等人研究 （2001）)使用连续发酵过程和在温石棉的固定化细胞,达到产乙醇收率约为97.3%。同样的系统进行自由细胞达到产率不高于84.5%。（Najafpour 等人研究 （2004））研究了海藻酸钙固定化酵母反应器中的酒精发酵。已经建议如琼脂、卡拉胶、聚丙烯酰胺凝胶、氧化淀粉、基西尔矿物和丝瓜海绵作为固定化酵母的载体(Holcberg和Margalith 1981;Elisabetta 1992; 989年,Ogbonna  2001等人研究）。

     1984年， Nagashima 等人研究海藻酸钙凝胶固定化酵母。固定化细胞应用于蔗糖流化床反应器乙醇连续发酵。一个总柱4000 L的试验工厂用来测试这个过程。它每天能够生产2400 L 乙醇。醇的生产力估算为每小时每升反应器体积为19.75 g ，转化率为95%，与之相比在自然环境中分离的呈现絮状能力酵母株的最大理论产量是8.5%(v / v)(Sree 2000;Nahvi 等人研究（2002）)。允许絮凝能力的酵母在固定化系统成为乙醇生产商。酵母絮凝已经被定义为“在酵母细胞中悬浮介质迅速形成块状和沉积物的现象。（Stewart 等人研究1976) 因此,它可能是絮凝细胞不需要惰性载体来维持反应器。

    1996年Stewart等人研究，通过使用摇床反应器与细胞循环倾注系统达到了每小时每升14.4 g乙醇的生产力和88.3%的收益率。2002年维埃加斯等人研究，使用两个无循环串联塔反应器获得每小时每升15.4 g的生产力。2003年维埃加斯研究了一个包括三个反应器的酵母菌株絮凝特性性能的系统。作者认为由于使用酵母菌株才获得每小时每升27.5 g 的高收益。尽管实验室中使用絮凝酵母研究结果流程是很有希望的,单特别需注意工业规模性的使用这种方法。这是由于事实上,即底物发酵(甘蔗汁或糖蜜)包含一种天然原生微生物群落，其可能主导整个过程,从而消除絮状酵母作为接种体。

    直到20世纪90年代中期,大多数巴西发酵单位开始于大量酵母的传统面包店行业。这个策略允许快速和安全启动,减少可能出现的问题与发酵事故。从20世纪90年代开始,酵母作为接种体被发现，并完全取代在这之前的原生酵母。在这个过程中之前仅有的相同单元中分离出的酵母能够留住。(Basso 等人研究, 1993）

    基于这一假设,这个工作的目的是为了验证使用不需要细胞分离和循环单元的乙醇生产过程中酵母菌株絮凝特性的可能性。酵母的絮凝特性允许自身固定在反应器。

2 方法

2.1 酵母菌株的筛选

分离出的酵母在沃勒斯坦实验室的营养琼脂上32摄氏度生长(WL营养培养基,DIFCO0424-17-9)7天 ，培养基保持在4摄氏度PDA(马铃薯葡萄糖琼脂- DIFCO 0549 -17 9)覆盖无菌矿物油。

在1995年- 2002年中分离出的酵母菌株在发酵还原环境中，下面的巴西乙醇生产单位:Usina戴安娜-直辖市Avanhadava - 圣保罗州(酵母11和酵母12 - 1995);Usina科斯塔平托-市(酵母6 - 1999和酵母9 -1998);Jales Usina Machado -直辖市(酵母10 - 1998);Usina琼奇拉——直辖市(酿酒- 1997,1999,酵母1、酵母2、酵母5 - 2000);中央德一个卢斯直辖市(酵母8- 2002)。

2.2 应变选择

    12株属于酵母属被选是由于它能在发酵罐环境中分离出来,因此适合工业使用,也是因为它们的絮凝能力增长方式。在视觉上进行对各菌株的絮凝能力的观察。12株种植在一个培养基中培养。这种类型的栽培允许视觉观察表明是否酵母能够以絮凝的方式生长。这种增长是形成团块分散在介质中,而不是在一个均匀介质。没有进行测试,以确定各种类型絮凝菌株的数量测试。

2.3 反应器使用

    这个系统是构建模拟工业操作条件。它包括两套玻璃反应器其容积1.4升,直径5厘米,高度65厘米。反应器串联、通过第一个反应器底部的蠕动泵帮助输送底物。发酵的酒离开了第一反应器顶部和滞留在第二个反应器的底部。气体通过位于第一反应器的顶部的出口排出。补料过程起源于一个稀释系统。浓缩的底物溶液被预期浓缩的水有序的稀释。所有试验在非无菌底物温度摄氏度32±0.5 的条件下进行(图1)。

2.4 介质的使用

    为了避免污染,进料的媒介反应器的进料介质有序的配备了一种浓缩液。这是在稀释系统中完成的，其中浓缩介质被水稀释的浓度描述如下。用于一号反应器进料介质组成如下:

    160 g / L蔗糖,6.0 g / L酵母提取物,5.0 g / L磷酸二氢钾,

    1.5 g / L氯化铵,1.0 g / L七水合硫酸镁和pH值5.0±0.2。

2.5。 接种物制备

    使用的菌株生长在PDA倾斜面，30摄氏度24小时.菌株悬浮于无菌水并转移到含有90毫升的合成培养基250毫升玻璃锥形瓶。锥形瓶是在150转、32摄氏度的摇床上培养24小时。12菌株混合为了形成一个接种体,并转让给反应器。使用稀释液1:500，在纽鲍尔(BOECO-Germany)改进的计数室细胞浓度得以确定。

2.6 试验取样

    每个反应堆获得接种量50%(约600毫升)。进料的生长底始于第一个反应器的底部,以获得3.75 h滞留时间。在这些操作条件下这个系统维持为15天。

 

    取样点:每日样本收集在四个不同的点:进料,反应器,反应器1和反应器2的基质以及反应器出口2。进行的分析是:进料板的总还原糖,第一反应器和第二反应器的总还原糖和乙醇和两座反应器中的酵母数量和分类。获得以下参数是每日核对这些数据:当前的酵母数量、产量水平(大规模生产乙醇的总还原糖质量表示在一个比例的最大理论产量(100·乙醇(g / L)/ 0.511（g/L))基板进料和转化量(底物消耗作为总可用基板的一个应变量）。

统计分析:一式三份样品分析总还原糖类乙醇和酵母数。描述性统计使用5.0、95%阿尔法误差的程序是用来确定的限制的手段信心。

2.7 床层高度

每日通过用一把尺直接测量,以确定细胞层高度。

2.8 解析方法

    在下列条件下，高效液相色谱法测定蔗糖、葡萄糖、果糖和乙醇的含量(高效液相色谱)。折光率检测器,87 h matacarb柱和0.1 N硫酸洗脱液。通过加入葡萄糖得到总还原糖,通过0.95的糖浓度的分工减少果糖和蔗糖浓度转化。

    定量的酵母菌在沃勒斯坦实验室营养培养基上通过扩散板技术，在32摄氏度条件下卵育出来。

 

2.9 鉴别存在于样品酵母菌株

12株菌株都显示合适发酵工业使用的性能。性能是根据发酵潜在技术评估的，(Andrietta et al。,1995)基于动力学参数(最大比生长速率)和具体的生产参数(乙醇和细胞收益率)。具体相对于有平均性能的参考酵母菌株其产品值已经达到。

    12酵母菌株一直以前描述的为:(一)在7天32摄氏度生长菌落形态为和(b)使用脉冲电场做核型凝胶电泳实验(但是脉冲场凝胶电泳的出现)。具体描述如下。

    Blondin and Vezinhet描述通过改进的方法，酵母染色体被分离。染色体在一个CHEF III系统中被分离，这种凝胶用溴化乙锭染色，以0.5 ll/ml预备在TAFE缓冲液中40 min，并又把TAFE缓冲液脱色半小时。在紫外线光中进行阅读。图像通过了BioImagem UVP系统分析。

     对每一个样品不同生物型的两个群落每天进行电泳核型分析。

 

 

 

 

 

 

 

 

 

 

 

 

 

Study of flocculent yeast performance in tower reactors for

bioethanol production in a continuous fermentation process with

no cell recycling

Abstract

The purpose of this study was to assess the retention ability of 12 different Saccharomyces sp. yeast strains with flocculent character-istics when inoculated in a continuous ethanol fermentation process. The system was comprised of two reactors connected in series with no cell recycling. The feeding substrate used was a synthetic medium containing glucose. The parameters assessed were total reducing sugars of the feeding substrate, total reducing sugars and ethanol at the outlet of the first and second reactors and quantification and classification of yeast population in the two reactors. The system reached yield levels of 83.53% of theoretical yield with a maximum total reducing sugars conversion of 92.68%. The conversion in this system was lower than expected. The dominant yeast in the process in both reactors, contrary to expectation, was the Saccharomyces CP6 strain which was unable to form pellets in spite of its flocculate

growth.? 2007 Published by Elsevier Ltd.

Keywords: Flocculation; Yeast; Biofuel; Ethanol; Fermentation

1. Introduction

The use of biofuel to replace oil is the most viable way

to ensure a sustainable future. The oil crisis in the 1970s

led Brazil to develop a national alcohol program (Pro ál-

cool) aimed at replacing gasoline for an alternative fuel.

Upon the introduction of Pro álcool, researchers were

encouraged to develop projects aimed at decreasing the

production cost of this biofuel. The pursuit of cost reduc-

tion of this biofuel has led to the development of alterna-

tive processes. Among these processes is the use of fixed

bed tower reactors using yeast strains with flocculent

characteristics. This type of process, when operated prop-

erly, may reach a level of high productivity preserving the

same yield rates obtained in conventional processes.

Recently, more attention has been given to immobilized

biomass activity, since such process has been acknowledged to play a significant role in bioreactor performance (Gikas and Livingston, 1997).

Numerous studies have suggested the production of ethanol with the use of immobilized yeasts. Joekes et al. (1998) immobilized the Saccharomyces cerevisiae strain onto chrysotile (Mg4Æ Si4O10(OH)3) The final yield was 26% higher than that with free cells. Wendhausen et al. (2001),working with continuous fermentation processes and immobilized cells in chrysotile, reached yield in ethanol of about 97.3%. This same system conducted as free cells reached yield not higher than 84.5%. Najafpour et al.(2004) studied alcoholic fermentation in a cell reactor of S. cerevisiae immobilized in calcium alginate. Substances such as agar, carrageenas, alginate, polycrylamide gels, oxystarch, mineral Kissiri and loofa sponge have been sug-

gested as support for immobilization of yeasts (Holcberg

and Margalith, 1981; Elisabetta et al., 1992; Kana et al.,

1989; Ogbonna et al., 2001).

Nagashima et al. (1984) immobilized S. cerevisiae in

calcium alginate gels. The immobilized cells were used in

fluidized-bed reactor for continuous ethanol fermentation

from sugar cane molasses. A pilot plant with a total col-

umn of 4000 L was constructed to test this process. It is

able to produce 2400 L/ethanol per day. The productivity

of ethanol was calculated as 19.75 g ethanol/L reactor vol-

ume hour with a 95% conversion yield versus the maximum

theoretical yield for the case of 8.5% (v/v) ethanol broth.

S. cerevisiae strains presenting flocculent ability have

been isolated in natural environments (Sree et al., 2000;

Nahvi et al., 2002). Flocculation ability of the S. cerevisiae

strains have allowed them to be regarded as ethanol pro-

ducers in autoimmobilized systems. Yeast flocculation

has been defined as ‘‘the phenomenon wherein yeast cells

adhere as clumps and sediment rapidly from the medium

in which they are suspended (Stewart et al., 1976). There-

fore, it is possible that flocculate cells remain within the

reactor with no need for inert support.

Paiva et al. (1996) reached productivities of 14.4 g of

ethanol/L h and yields of 88.3% by using a system of fluid-

ized bed reactors with cell recycling through decantation.

Viegas et al. (2002) used two tower reactors connected in

series with no cell recycling and obtained productivity of

15.4 g/L h. Viegas (2003) studied the performance of a

yeast strain with flocculation characteristics in a system

comprising three reactors in series. This author attributed

the high yield obtained (27.5 g/L h) to the yeast strain used.

Even though the laboratory-scale results of the studies of

processes using flocculent yeasts are quite promising, spe-

cial attention should be given to the use of this approach

on an industrial scale. This is due to the fact that the sub-

strate to be fermented (sugarcane juice or molasses) con-

tains a flora of native microorganisms which may

dominate the process and consequently eliminate the floc-

culent yeast used as inoculum.

Until the mid 1990s, most Brazilian fermentation units

traditionally started the season with tons of yeast from

the bakery industry. This strategy allowed a fast and safer

start-up, minimizing possible problems with fermentative

accidents. From the 1990s on, yeasts used as inoculum were

found to be completely replaced by native yeasts at the

beginning of the season. The only yeast able to remain in

the process was the one of the same unit isolated in previ-

ous seasons (Basso et al., 1993).

Based on this hypothesis, the purpose of this work was

to verify the possibility of using yeast strains with floccu-

lent characteristics isolated from industrial processes in

an ethanol production process which does not require the

cell separation and recycling unit. The flocculation charac-

teristic of the yeast allows it to be immobilized inside the

reactor.

2. Methods

2.1. Yeast strain isolation

The isolated yeasts were grown in the Wallerstein Labo-

ratory Nutrient Agar (WL nutrient medium – DIFCO0424-17-9) for 7 days at 32 ?C. Cultures were maintained

at 4 ?C on PDA (Potato Dextrose Agar – DIFCO 0549-17-9) covered with sterile mineral oil.

Yeast strains were isolated during the years 1995–2002

in fermentation vat environments of the following Brazil-

ian ethanol producing units: Usina DIANA – municipality

of Avanhadava – State of Sa ˜o Paulo (Saccharomyces4,

Saccharomyces11 and Saccharomyces12 – 1995); Usina Costa Pinto – Municipality of Piracicaba – State of Sa ˜oPaulo (Saccharomyces6 – 1999 and Saccharomyces9 –1998); Usina Jales Machado – municipality of Goiane ˆsia– State of Goia ´s (Saccharomyces10 – 1998); Usina Junqueira – municipality of Igarapava – State of Sa ˜o Paulo (Saccharomyces – 1997, Saccharomyces2 1999, Saccharomyces1 and Saccharomyces5 – 2000); Central de A´lcool Luce ´lia – municipality of Luce ´lia – State of Sa ˜o Paulo (Saccharomyces8 – 2002).

2.2. Strain selection

The 12 strains belonging to the Saccharomyces genus

were chosen for their ability of being isolated in fermenta-

tion tanks environments, thus suitable for industrial use,

and for their ability to grow in a flocculate manner. The

observation of the ability of each of the strains to flocculate

was carried out visually. The 12 strains were cultivated in a

liquid medium. This type of cultivation allowed visual

observation to indicate whether the yeast is able to grow

in a flocculate manner. This growth is explained by the for-

mation of clumps dispersed in the medium rather than in a

homogeneous medium. No tests to quantify the type of

flocculation of each strain were carried out.

2.3. Reactor used

This system was constructed in order to simulate indus-

trial operation conditions. It comprises two jacket glass

reactors with 1.4 L volume, 5 cm diameter, and 65 cm

height. The reactors were connected in series, with sub-

strate feeding by the help of a peristaltic pump in the bot-

tom of the first reactor. The fermented wine exited from the

top of the first reactor and continued to the bottom of the

second reactor. Gases of the first reactor were eliminated

through an outlet located at the top. Substrate feeding

the process originated from a dilution system. The concen-

trated substrate solution was diluted in line with water for

the intended concentration. All trials were conducted with

non-sterile substrate at 32 ± 0.5 ?C (Fig. 1).

2.4. Medium used

In order to avoid contamination, the feeding medium of

the reactor was prepared in line from a concentrated solu-

tion. This was done in the dilution system, wherein the

concentratedmediumwasdilutedwithwaterfortheconcen-

tration described hereafter. The medium used for feeding

reactor 1 was synthetic with the following composition:

160 g/L sucrose, 6.0 g/L yeast extract, 5.0 g/L monobasic

potassium phosphate, 1.5 g/L ammonium chloride, 1.0 g/L

hepta hydrated magnesium sulphate and 1.0 g/L potassium

chloride pH 5.0 ± 0.2.

2.5. Inoculum preparation

The strains used were individually grown in PDA slant

for 24 h at 30 ?C. The strains were suspended with sterile

water and transferred to 250 ml Erlenmeyer flasks with

90 mL of synthetic medium. Flasks were incubated in a

shaking incubator for 24 h at 32 ?C and 150 rpm. The 12

strains were mixed in order to form a single inoculum,

which was transferred to the reactors. Cell concentration

was determined in a Neubauer (BOECO-Germany) improved counting chamber, using a dilution of 1:500.

2.6. Trial sampling

Each reactor received 50% inoculum (about 600 ml each). Feeding of the growing medium was started from the bottom of the first reactor so as to obtain 3.75 h residence time. This system was kept under these operation conditions for 15 days. Sampled points: Samples were collected daily in four different points: feeding substrate, reactor 1, reactor 2 and outlet of reactor 2. The analyses carried out were: total reducing sugars of the feeding substrate, total reducing sugars and ethanol of the first reactor and outlet of secondreactor and quantification and classification of yeast population (CFU/ml) in the two reactors. The following para-meters for each trial day were obtained with these data:amount of each of the yeasts present, yield levels (ethanol mass produced by total reducing sugars mass expressed

on a percentage of maximum theoretical yield (100· etha-

nol (g/L) /0.511 substrate in the feed (g/L)) and conversion

(substrate consumed as a function of the total available

substrate).Statistical analysis: Samples to analyze total reducingsugars ethanol and yeast population were drawn in triplicate. Descriptive statistics from Statistic 5.0 program using 95% of alpha error was used to determine the confidence

limits for the means  

.2.7. Bed height

 The cells bed height was determined daily through direct measurement, using a ruler.

2.8. Analytic methodology

 Sucrose, glucose, fructose and ethanol determination was carried out by High pressure liquid chromatography (HPLC) in the following conditions: refraction rate detector, 87H matacarb column and as 0.1 N sulphuric acid eluent. Total reducing sugar was obtained by adding glucose,fructose and sucrose concentration converted in reducing sugar by the division of its concentration for 0.95.

Quantification of yeasts was carried out with the spread

plate technique in Wallerstein laboratory nutrient (WLN)

growing medium. Plates were incubated for 7 days at 32 ?C

2.9. Identification of yeast strains present in samples

All 12 strains have shown appropriate fermentative

performance for industrial use. The performance was

assessed according to the fermentative potential technique

(Andrietta et al., 1995) based on the determination of the

kinetic parameter (maximum specific growth speed) and

specific production parameters (ethanol and cellular

yields). Specific productions values have been compared

to the ones reached by a reference Saccharomyces strain,

which has average behavior regarding the strains used in

Brazilian distilleries.

The 12 yeast strains had been previously characterized

as: (a) colony morphology when grown in WLN for 7 days

at 32 ?C, and (b) electrophorectic karyotype using pulsed-

field gel electrophoresis (PFGE) described as follows.

Yeast chromosomes were isolated by a modified method

described by Blondin and Vezinhet (1988). The chromo-

somes were separated in a CHEF III?system (Bio-Rad)

The gel was colored with ethidium bromide prepared in

TAFE buffer at 0.5 ll/ml for 40 min and discolored with

TAFE buffer for half an hours. Reading was carried out

in ultra violet light. Images were analyzed through the

UVP BioImagem System.

Electrophoretic karyotyping has been carried out for the

two colonies of each one of the different biotypes present in

the sample in each trial day.