Journal of Biorheology

, Volume 24, Issue 1, pp 29–35

Construction and application a vane system in a rotational rheometer for determination of the rheological properties of Monascus ruber CCT 3802

Authors

  • F. Vendruscolo
    • Chemical and Food Engineering DepartmentFederal University of Santa Catarina, UFSC
  • L. O. Pitol
    • Chemical and Food Engineering DepartmentFederal University of Santa Catarina, UFSC
  • B. A. M. Carciofi
    • Chemical and Food Engineering DepartmentFederal University of Santa Catarina, UFSC
  • D. E. Moritz
    • Chemical and Food Engineering DepartmentFederal University of Santa Catarina, UFSC
  • J. B. Laurindo
    • Chemical and Food Engineering DepartmentFederal University of Santa Catarina, UFSC
  • W. Schmidell
    • Chemical and Food Engineering DepartmentFederal University of Santa Catarina, UFSC
    • Chemical and Food Engineering DepartmentFederal University of Santa Catarina, UFSC
Original Article

DOI: 10.1007/s12573-010-0019-7

Cite this article as:
Vendruscolo, F., Pitol, L.O., Carciofi, B.A.M. et al. J Biorheol (2010) 24: 29. doi:10.1007/s12573-010-0019-7

Abstract

The objectives of this work were the construction and adaptation of a vane system in a rotational rheometer; calibration of vane system using a mineral oil, and determination of rheological properties of the broth fermentation of Monascus ruber CCT 3802 at high cell density. Batch fermentation was carried out with glucose as the sole carbon source in a Bioflo III bioreactor. The consistency index (K) and flow behavior index (n) of the broth fermentation of M. ruber were characterized at different biomass concentrations and were adequately described by a power law model. The K and n values were significantly affected by biomass concentrations. K values ranged from 0.375 to 11.002 Pa sn when evaluated at biomass concentrations from 25.67 to 63.20 g L−1. The pseudoplastic behavior was confirmed by values of n that ranged between 0.157 and 0.254. Simple empirical correlations have been proposed to quantify the dependence of the power law terms on fungal biomass concentration.

Keywords

RheologySubmerged fermentationMass transferHigh cell densityVane systemMonascus ruber

Introduction

The metabolic performance of a microbial culture in a bioreactor depends markedly on complex interactions between the various operating conditions. The stirring intensity, the microbial species being cultured, and the type and amount of added nutrients determine the bulk rheology and cell morphology [1]. Knowledge of the rheological properties of a fermentation broth is very important for designing the operating conditions, for example stirring and aeration [2]. It is also important to know how the composition of the fermentation broth (biomass and nutrients concentrations) affects its rheological behavior.

The specific growth morphology produced under given conditions depends on several factors including the fungal strain, the initiation culture (e.g. spores, pellets, and dispersed mycelium), the nature of the medium and the hydrodynamic regime in the bioreactor [1]. The flow pattern of fermentation broths containing filamentous fungi differs substantially from that of bacterial cultures, because of the complex morphology of filamentous fungi [3]. The rheological behavior is closely related to the morphology and biomass concentration [4]. The broth rheology affects the transport phenomena in bioreactors and must be considered to order for improve the yield of a desired product [5].

Generally, plant cell broths exhibit a pseudoplastic behavior, which depends on biomass concentration, cell aggregation and cell morphology [6]. The biomass concentration has been considered to be the main factor affecting this pseudoplastic behavior of broths. Free filaments have usually been described by properties such as length of the main hyphae, number of hyphal tips, total hyphae tips, and main hyphal thickness [7]. Mycelial fermentation broths are initially Newtonian because of the low biomass concentration. However, during the fermentation there is an increase in biomass concentration, which considerably alters its rheology [3].

The effect of biomass concentration on broth rheology is a complex problem because microbial morphology greatly affects the broth flow behavior. As the morphology of the filamentous microorganism is affected by a variety of factors, for example stirring conditions (type and speed) and dissolved oxygen concentration, it is difficult to relate the broth rheology to every aspect of microbial morphology [8].

An important feature of a batch fermentation process is the change in rheological behavior during fermentation because of the variation of substrate composition, biomass concentration, and morphology of the microorganisms. Generally, two growth forms, the filamentous and the pelleted forms can be observed in most fungal fermentations; the pelleted form is usually less viscous than the filamentous form [9]. Also, at high microbial concentrations, the hyphae can become entangled, resulting in a highly viscous suspension and possibly leading to very non-Newtonian behavior. Previous studies have shown that fermentation broths containing high concentrations of filamentous microorganisms are highly viscous and are characterized by shear rate-dependent viscosities and by a yield stress [3, 10].

Submerged cultures of filamentous fungi are widely used to produce commercially important metabolites including many antibiotics, enzymes, pigments, and other products. In a submerged fermentation, Monascusruber can grow in a variety of morphological forms, varying between a network of freely dispersed mycelia to tightly packed, discrete pellets. The rapid increase in viscosity that accompanies filamentous growth greatly reduces oxygen transfer. The available literature shows the effects of mechanical forces on fungal morphology and its rheological properties consistency index (K) and flow behavior index (n) are dependent on biomass concentration [1114] although there are no specific data on Monascusruber fermentations.

Biological materials, for example many foods and broth fermentations, are suspensions of solid matter in a continuous medium. In addition, a thin layer of the continuous phase may separate at the moving and stationary surfaces, and this can induce errors in the measured values of shear rate [15, 16]. For these and other complex material, approximate shear rate–shear stress data can be obtained by a technique using a rotating vane that minimizes settling and separation of the product [15].

However, the use of rotational rheometer to determine the rheological properties of viscous mycelial suspensions is often unsatisfactory. The objectives of this work were the construction and application of a vane system as a special spindle in a rotational rheometer to determine rheological properties of the broth fermentation of Monascus ruber CCT 3802. The technique was developed for the determination of power consumption during mixing of fluids and adapted to study of the rheology of fermentation broths.

Materials and methods

Microorganism and culture media

Monascus ruber CCT 3802 was obtained from the Tropical Culture Collection André Tosello (Campinas—SP, Brazil). The stock culture was maintained on potato dextrose agar (PDA) tubes. Tubes and Roux bottles were inoculated, incubated at 30°C for 7 days and subsequently stored at 4°C. The inoculum culture medium contained, per liter: 20 g glucose, 5 g glycine, 2.5 g KH2PO4, 2.5 g K2HPO4, 0.5 g MgSO4·7H2O, 0.1 g FeSO4·7H2O, 0.1 g CaCl2, 0.03 g MnSO4, and 0.01 g ZnSO4. The initial pH was adjusted to 4.0 before sterilization. The fermentation culture medium contained, per liter: 20 g glucose, 5 g glycine, 2.5 g KH2PO4, 2.5 g K2HPO4, 0.5 g MgSO4·7H2O, 0.1 g FeSO4·7H2O, 0.1 g CaCl2, 0.03 g MnSO4, and 0.01 g ZnSO4. The initial pH was adjusted to 3.0 after sterilization.

Inoculum preparation

Monascus ruber CCT 3802 was initially grown on PDA medium in a Roux bottle incubated at 30°C for 7 days and subsequently stored at 4°C. A suspension of spores was obtained by washing the Roux bottle cultures with a sterile aqueous solution of 0.1% of Tween 80. Fungal pellets were obtained by germination of spores suspended in a 1,000 mL shaken baffled flask containing 400 mL inoculum culture medium at 30°C on rotary shaker at 120 rpm for 60 h, and used for further inoculation in a bioreactor.

Stirred tank fermentation

Fermentation was carried out in a 6-L batch bioreactor (Bioflo III from New Brunswick Scientific Co., New Jersey, USA) for 4 days (working volume of 4 L) with a vessel with internal diameter of 0.170 m, a rounded bottom and a height/diameter ratio of 1.4. Agitation was provided by two six-blade Rushton turbines with a D/T ratio of 0.38 and a W/D ratio of 0.18. The culture medium (3.6 L) was inoculated with 0.4 L of the inoculum culture (10% v/v; = 0.5 g L−1 dry equivalent of cells), under the conditions: temperature 30°C; stirring speed 300 rpm; specific aeration rate 0.6 vvm, and initial pH 3.0.

Biomass determination

Dry weight was determined gravimetrically by filtration of 10-mL samples through a Whatman filter paper no. 1 (Madiston, UK). The 100 pellet diameters were obtained using a microscope Bioval (Warszawa, Poland) with a charge-coupled device camera Alder CCTV (Philippines).

Rheological measurements

Rheological measurements were performed in a rheometer Rheotest 2.1 (MLW, Prüfgeräte-Werk, Germany) equipped with a circulating thermostatic bath (model MQBMP-01, Microquímica, Florianópolis, Brazil) to control the working temperature. The measurements were carried out using a standard concentric cylinders spindle system and a vane spindle system, sketched in Fig. 1.
https://static-content.springer.com/image/art%3A10.1007%2Fs12573-010-0019-7/MediaObjects/12573_2010_19_Fig1_HTML.gif
Fig. 1

Rotational vane system used with the rotational rheometer (D = 0.037 m, H = 0.040 m, l = 0.010 m, L = 0.002 m and Ht = 0.18 m)

The instrument can be operated at 44 different speeds, which are changed stepwise with a selector switch. The speed of the rotating spindle (N) varied from 0.028 to 243 rpm.

Rheological fluids

Rheological tests were performed with three different fluids: a standard oil (IPT 86 standard oil, Instituto de Pesquisas Técnológicas, São Paulo, Brazil) as a Newtonian fluid with well known density (ρ) and viscosity (μ); guar gum (Sigma–Aldrich, St. Louis, USA) aqueous solution (1% w/v) as a non-Newtonian fluid; and the culture broth of Monascus ruber, the objective of this work. Fluids were maintained at 30°C in all tests.

Rheometer rheotest

Operation
In order to determine the flow curve of a fluid using a rotational rheometer system, the average shear rate (γav) and the average shear stress (τav) at different N values must be determined. The impeller torque (T) as a function of N was measured using rotational spindle systems (vane or concentric cylinders). According to Metzner and Otto [17], for a given fluid, the γav and the τav should be determined by Eqs. 1 and 2, respectively.
$$ \gamma_{\text{av}} = aN $$
(1)
$$ \tau_{\text{av}} = bT $$
(2)
where a and b are calibration constants which depend exclusively on system geometry.

a and b for standard concentric cylinder systems are provided by the equipment’s manufacturer. However, for the vane spindle system, they may be determined experimentally.

Vane spindle system calibration

The calibration procedure employed was based on the original reports of Bongenaar et al. [18], slightly modified by Kemblowski and Kristiansen [19] and by Badino Jr. et al. [16].

The impeller power number (NP) and the Reynolds number (Re) for a mixing device operating in laminar flow are related by Eq. 3 [20].
$$ N_{P} = {\frac{c}{Re}}$$
(3)
where the constant c (shape factor) depends on the specific geometry of the mixing system.
The generalized Re and NP numbers are defined by Eqs. 4 and 5 [20], respectively.
$$ {Re} = {\frac{{\rho ND^{2} }}{{\mu_{\text{ap}} }}} $$
(4)
$$ N_{P} = {\frac{P}{{\rho N^{3} D^{5} }}} $$
(5)
where ρ is the fluid density, D is the cylinder or vane spindle diameter (as showed in Fig. 1), μap is the apparent viscosity and P is the stirring power consumption, defined as P = 2πNT.
Substitution of Eqs. 4 and 5 in Eq. 3 leads to Eq. 6:
$$ c = {\frac{2\pi T}{{\mu_{\text{ap}} D^{3} N}}} $$
(6)
The average shear stress of a fluid was characterized by the power law model of Ostwald–de Waele (Eq. 7) [15].
$$ \tau_{\text{av}} = K\gamma_{\text{av}}^{n} $$
(7)
where, the constants K and n represent the consistency index and the flow behavior index, respectively.
Equation 8 defines the average shear stress as an apparent viscosity and average shear rate function. Then, the apparent viscosity is given by Eq. 9.
$$ \tau_{\text{av}} = \mu_{\text{ap}} \gamma_{\text{av}} $$
(8)
$$ \mu_{\text{ap}} = K\gamma_{\text{av}}^{n - 1} $$
(9)
Combining Eqs. 6 and 9, the γav could be determined by use of Eq. 10 [21].
$$ \gamma_{\text{av}} = \left( {{\frac{{cKD^{3} N}}{2\pi T}}} \right)^{{\left( {{\frac{1}{1 - n}}} \right)}} $$
(10)
The slope of the γav versus N plot is the calibration constant a, determined by linear regression of Eq. 1 for the vane impeller results. The calibration constant b was calculated using Eq. 11, obtained from Eqs. 1, 2, 6 and 8.
$$ b = {\frac{2\pi }{{cD^{3} }}}a $$
(11)

Broth rheology

The consistency index behavior of broth rheology was expressed as a function of biomass concentration, as can be seen by Eq.  12 [22, 23].
$$ K = K_{0} X^{m} $$
(12)
where K0 is a pre-exponential factor, X is a biomass concentration, and m is an exponential factor.

Results and discussion

Construction and calibration

A typical rheogram of standard IPT 86 oil, in the concentric cylinders system, is shown in the Fig. 2. It can be seen that the oil had Newtonian behavior (τav vs γav plot was highly linear, with R2 = 0.9997), as expected, with a viscosity of 1.032 Pa s, 1.4% less than the viscosity of IPT 86 oil at the same temperature (1.047 Pa s).
https://static-content.springer.com/image/art%3A10.1007%2Fs12573-010-0019-7/MediaObjects/12573_2010_19_Fig2_HTML.gif
Fig. 2

Rheogram of standard oil IPT 86 on the concentric cylinders system

The impeller torque (T) as a function of N, measured with the rotational spindle vane, is presented in Fig. 3. Values of Re for laminar region (Eq. 4) and NP (Eq. 5) (IPT 86 oil) for the vane system, are plotted in Fig. 4. NP versus (1/Re) data were highly linear (R2 = 0.9997), and the constant c value was determined from curve slope.
https://static-content.springer.com/image/art%3A10.1007%2Fs12573-010-0019-7/MediaObjects/12573_2010_19_Fig3_HTML.gif
Fig. 3

Torque versus n for standard IPT 86 oil stirred in the vane system

https://static-content.springer.com/image/art%3A10.1007%2Fs12573-010-0019-7/MediaObjects/12573_2010_19_Fig4_HTML.gif
Fig. 4

NP versus 1/Re plot in the region of laminar flow (standard oil IPT 86)

Figure 5 presents the rheogram of a 1% (w/v) guar gum suspension measured at 30°C in the concentric cylinders system. The power law (Eq. 7) was fit to the experimental γav versus τav data in order to determine K and n values for this suspension, 1,272.7 mPa sn and 0.257, respectively. These results are consistent with those obtained by Kayacier and Dogan [23], who determined the consistency index (2,283.2 mPa sn) and flow rate index (0.319) of 1% (w/v) guar gum suspension at 25°C.
https://static-content.springer.com/image/art%3A10.1007%2Fs12573-010-0019-7/MediaObjects/12573_2010_19_Fig5_HTML.gif
Fig. 5

Flow curve for 1% (w/v) guar gum aqueous solution at 30°C

The next step was performing experiments with the guar gum solution in the vane system. At each impeller speed N, the corresponding torques, T, were measured. From the constants c, K and n, previously determined for the vane system, it was possible to calculate the corresponding values of γav, using Eq. 10.

The value of the constant a was determined as the slope of the curve γav versus N (Fig. 6), according to Eq. 1. Once the value of a is known, it is possible to determine, using Eq. 11, the values of the second calibration constant (b). With both constants values, Eqs. 1 and 2 for the vane spindle system are given by Eqs. 13 and 14, respectively.
https://static-content.springer.com/image/art%3A10.1007%2Fs12573-010-0019-7/MediaObjects/12573_2010_19_Fig6_HTML.gif
Fig. 6

Determination of calibration constant a for the vane system

$$ \gamma_{\text{av}} = 14.28N $$
(13)
$$ \tau_{\text{av}} = 739.19T $$
(14)

Rheological behavior of the fermentation broth

By using the calibrated vane system, the rheological behavior of the fermentation broth at different biomass concentrations was determined. The biomass was characterized by measuring the diameters of the pellets. The pellet diameters were found to be in the range of 0.8–1.7 mm, 60% of the pellets being in the range of 1.4–1.7 mm (data not shown).

Figure 7 shows τ versus γ for broth fermentation at different biomass concentrations, analyzed in the vane system at 30°C. The experimental data were analyzed using Eq. 9 for its simplicity and application to a great variety of fluids and wide use in engineering calculations.
https://static-content.springer.com/image/art%3A10.1007%2Fs12573-010-0019-7/MediaObjects/12573_2010_19_Fig7_HTML.gif
Fig. 7

Curves obtained for Monascus ruber CCT 3802 suspensions using the vane system

Broth rheology of Monascus ruber cultures exhibited non-Newtonian behavior for all biomass concentrations (Fig. 7). The pseudoplastic behavior was confirmed by the values of n, which ranged between 0.157 and 0.254. The values of K ranged from 0.375 to 11.002 Pa sn. The non-Newtonian behavior is relatively predominant at lower shear rate and higher mycelial concentration. These phenomena may be because higher energy dissipation rates deform the three-dimensional structure of highly entangled mycelia and break down the pellets of the mycelial aggregate network, which is negligible at very high shear rates.

Rheological analysis of clarified broth samples, using the concentric cylinders rheometer at 30°C, showed a Newtonian viscosity of 1.8 mPa s (data not shown), indicating that the pseudoplasticity of the broth was entirely due to the biomass and not to the aqueous solution with extracellular metabolites. Mycelial fermentation broths are initially Newtonian because of the low biomass concentrations. However, during the fermentation, the increase of biomass concentration substantially changes the rheology of the fermentation broth to pseudoplastic behavior [24].

Figure 8 shows the behavior of the consistency index (K) and the flow behavior index (n) as a function of biomass concentration, obtained by applying the power law (Eq. 7) to the experimental data presented in Fig. 7.
https://static-content.springer.com/image/art%3A10.1007%2Fs12573-010-0019-7/MediaObjects/12573_2010_19_Fig8_HTML.gif
Fig. 8

Flow behavior index (n) and consistency index (K) as a function of biomass concentration (X)

Consistency index data (Fig. 8) were fit by a power law-type model (Eq. 7) resulting in K = 10−5X3.824 Pa sn. These results can be compared with the results obtained by Allen and Robinson [12] who studied Aspergillus niger (K = 4.3 × 10−4X3.3 Pa sn), Penicillium chrysogenum (K = 3.6 × 10−3X2.5 Pa sn), and Streptomyces levoris (K = 0.27 X0.7 Pa sn), and those obtained by Queiroz et al. [22] working with Aspergillus awamori (K = 0.89 X1.48 Pa sn). The different values can be explained by the predominant morphological characteristic exhibited by each type of microorganism, the presence of pellets, dispersed mycelia and agglomerates. It is important to remark that, in all cases, the fermentation broth had pseudoplastic behavior.

The flow index n varies little (0.157–0.254) during experiments performed at difference biomass concentrations because variation of pellet diameter is small. These results may be compared with that obtained by Pollard et al. [25], where n varied from 0.35 to 0.40 at biomass concentrations of 20–65 g L−1, respectively. This a typical behavior observed with filamentous fungi, and it is very well accepted that rheological properties of mycelial cultures are determined both by biomass and the morphological state of the operating conditions in the bioreactor [26].

The results obtained in this work highlight the importance of determining the effect of biomass concentration on the rheological behavior of culture media. Non-Newtonian behavior of plant cell suspension cultures causes several problems of mixing and oxygen transfer [16, 27, 28]. Obtaining the behavior of the consistency index as a function of the biomass concentration enables control of oxygen transfer in cultures of microorganisms, which is very important when dealing with processes of aerobiosis.

Conclusion

Use of a vane system in a rotational rheometer is a good means of determining the rheological behavior of fermentation broth with biomass and other solids dispersed. The calibration procedure is simple, and very important to obtaining reliable data on the rheological properties characterizing the fermentation broth. The fermentation broth of Monascusruber CCT 3802 has pseudoplastic behavior for all biomass concentrations investigated in this work, in agreement with literature results. In this way, the procedure and results presented in this work can be useful for rheological characterization of many kinds of fermentation broth and other kinds of suspensions.

Acknowledgments

The authors wish to thank the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq—The National Council for Scientific and Technological Development) for financial support (process number 476056/2006-3) and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES—Committee for Professional Development of Higher Education Staff) for a research scholarship.

Copyright information

© Japanese Society of Biorheology 2010