Elsevier

Polymer

Volume 43, Issue 8, April 2002, Pages 2375-2383
Polymer

Polymer melt rheology with high-pressure CO2 using a novel magnetically levitated sphere rheometer

https://doi.org/10.1016/S0032-3861(01)00804-7Get rights and content

Abstract

A magnetically levitated sphere rheometer (MLSR) designed to measure viscosity of fluids exposed to high-pressure carbon dioxide has been developed. This device consists of a magnetic sphere submerged inside a test fluid within a high-pressure housing and levitated at a fixed point. The housing is constructed from an optically transparent sapphire tube. The cylindrical tube can be moved vertically to generate a shear flow around the levitated sphere. The difference in magnetic force required to levitate the sphere at rest and under fluid motion can be directly related to fluid viscosity. Rheological properties, specifically zero shear viscosities, of transparent high-pressure materials can be measured to a precision of about 5% and over a wide range of viscosities. In addition, operation at constant pressure, in concentration regimes from a pure polymer to an equilibrated polymer/supercritical fluid solution, and at shear rates over several orders of magnitude is possible, eliminating many of the disadvantages associated with other high-pressure rheometers. Experiments performed at different temperatures with a poly(dimethylsiloxane) melt at atmospheric pressure are compared with data from a commercial Couette rheometer to demonstrate device sensitivity and viability. Measurements of a PDMS melt plasticized by high-pressure CO2 are performed to illustrate the utility of the new rheometer under high-pressure conditions. Experimental data are obtained at 30 °C, for pressures up to 20.7 MPa and CO2 concentrations reaching 30 wt%. Viscosity reductions of nearly two orders of magnitude compared with the pure polymer viscosity at atmospheric pressure are observed. Additionally, the effects of pressure on a polymer/CO2 system are directly investigated taking advantage of the constant pressure operation mode of the MLSR. This allows us, for the first time in experiments of polymers with supercritical fluids, to decouple the effects of CO2 concentration and pressure in a single device.

Introduction

In recent years, the promotion and use of environmentally friendly solvents in a variety of traditional chemical and engineering processes has received increased attention. Liquid and supercritical CO2 is one of the leading solvents promoted as an environmentally friendly or ‘green’ alternative to conventional organic solvents [1]. Unfortunately, implementation of supercritical fluids (SCFs) in industrial processes has been problematic. These problems often result from large knowledge gaps in the physical properties and processing characteristics of SCF/polymer systems. Most importantly the effect of CO2 and other SCFs on the rheological behavior of polymer melts and solutions have only been studied at very low or high concentrations of SCFs. In particular, minimal experimental data is available on viscosities of CO2 solutions, which is required for equipment design, industrial scale-up, and process modeling of new SCF applications. A number of experimental devices capable of measuring rheological properties of complex fluids under high-pressure have been developed over the last 40 years to address this lack of knowledge [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14]. Most of these devices can be classified into three categories: (a) pressure driven, (b) falling body, and (c) rotational devices. Each of these devices has specific advantages, which make them important techniques for rheological measurement. However, each also has significant disadvantages, which limits their usefulness, especially for CO2/polymer solutions.

Pressure driven devices such as back-pressure regulated capillary rheometers [2], [3], [15], [16], [17] and extrusion slit dies [4], [5], [6], [7], [8] can measure polymer melts plasticized by CO2 and other SCFs. These devices mimic conventional polymer processing giving much needed viscosity information for many polymer melts. However, the nature of a pressure driven flow limits their utility. In particular, the often large pressure drop limits the concentration of the dissolved SCF that can be measured. Therefore, the SCF concentration must be maintained at low values, far from equilibrium, to avoid phase separation during measurement. In this low SCF concentration region, these devices have been found to provide accurate and useful rheological information.

Falling body devices such as falling cylinder [9], [10], [11] or rolling sphere [18] devices are useful in measuring highly concentrated SCF solutions of low viscosity. They operate at constant pressure allowing direct measurement of pressure effects on rheology [11]. Unfortunately, the sinker density and fluid viscosity determine the shear rate, making the measurement of viscosity as a function of shear rate difficult [11]. Falling body devices are also limited to low-viscosity solutions because the fall times required to measure high viscosity systems are too long to be experimentally viable. Additionally, limited shear rate control makes data comparison between different experimental samples difficult.

Finally, rotational devices such as high-pressure couette [14] or parallel plate geometries [12], [13] can operate at constant pressure and a variety of different shear rates. Unfortunately, rotational devices are difficult to design because viscoelastic or rotational information must be transferred under pressure. Mechanical information is often lost via transfer of torque through a dynamic seal, limiting measurement sensitivity. Additionally, a magnetically coupled drive shaft, which generates slightly non-uniform flow, is often used for motion control. These devices also generally require operation to occur at or near equilibrium conditions because a headspace of the pressurizing SCF must be maintained. The polymeric sample can often leak out of the measurement device and penetrate the SCF headspace due to polymer swelling, creating measurement errors.

Our study focuses on the development of a new class of rheometer for measuring the properties of SCF/polymer solutions and melts, the magnetically levitated sphere rheometer (MLSR). This device is based on the original designs of an ambient pressure MLSR proposed by Adam and Delsanti [19] and later modified by others, [20], [21], [22] used to measure the rheological properties of polymer gels and volatile polymer solutions [19], [23], [24], [25], [26], [27]. The principles and device designs are also similar to low pressure magnetic float densimeters [28], [29], [30], [31], [32]. The high-pressure MLSR (shown schematically in Fig. 1) employs the basic cell design of a falling sphere rheometer, allowing the device to operate at constant pressure conditions and eliminating the headspace associated with the high-pressure rotational devices. In contrast to the falling body devices, however, the sphere is held stationary at a fixed point through magnetic levitation while the cylindrical sample chamber is moved vertically using a stepper motor to generate shear flow around the stationary sphere. The sample chamber velocity is coupled to the shear rate and the change in magnetic force necessary to maintain the sphere stationary is related to the shear stress, thus allowing measurement of steady shear viscosity. However, the device needs to be calibrated against a known fluid viscosity because of the non-homogeneous nature of the flow.

The device is not limited to low-viscosity materials, as in the falling body devices, because measurement time is only proportional to the response of the instrument's electronic control and not the viscosity of the system. The MLSR uses the magnetic force to detect the shear stress imposed on the sphere by the sample chamber motion, eliminating the error of magnetically coupled motion associated with many high-pressure rotational devices. These advantages make the MLSR an excellent candidate for high-pressure rheometry because both the application of shear and the measurement of shear stress can be accomplished without transferring information directly across a mechanical seal.

In this paper, we discuss design aspects of the novel high-pressure magnetically levitated sphere rheometer (MLSR). The fabrication of both the MLSR and the high-pressure sample cell are examined in detail, together with calibration of the device. A comparison of data obtained using the MLSR and a commercially available rheometer at atmospheric pressure is shown to establish the accuracy of the new experimental device. Finally, we present results of a set of high-pressure experiments to demonstrate the utility of this device under pressurized conditions, for the measurement of zero shear viscosity of Newtonian fluids and use this data to examine some of the assumptions of previous experimental work with relation to pressure and CO2 concentration.

Section snippets

Sphere levitation

To describe the operation of the MLSR [19], only simple force balances about the magnetically levitated sphere are required. For simplicity, a magnetic sphere of radius a is considered to be maintained at the origin of a three-dimensional coordinate system. Levitation of the sphere is accomplished by balancing the effective gravitational force on the sphere by an applied magnetic force from an exterior coil, as described by Eq. (1):FMo=Fg−Fbwhere FMo is the magnetic force required to levitate

Rheometer design and construction

A schematic diagram of the MLSR is shown in Fig. 1. The main housing or temperature bath, based on the original designs of Adam and Delsanti [19], were machined in-house using non-magnetic stainless steel. The main housing is used for temperature control of the rheometer. Using ports on both the top and bottom of the main housing, cooling water, filtered for particles that would create noise in the optical detection system, is circulated around the sample chamber. The control of the sample

Rheometer calibration and experimental verification

Because the ultimate reason for constructing this device is to measure viscosities of polymer melts with dissolved liquid and supercritical CO2, a set of viscosity standards with a range of viscosities in the area of interest were chosen to calibrate the instrument. While the calibration is expected to be linear over a large range of viscosities the measurement of low viscosity material requires enhanced measurement sensitivity. This can be accomplished by an appropriate choice of magnetic

Viscosity measurements

To demonstrate the high-pressure capability of the MLSR, the rheological effects of CO2 incorporation into a polymer were measured. A known mass of the sample PDMS (Viscal-100M) was loaded into the sample cell. A vacuum was then pulled on the sample to remove any air dissolved in the system followed by subsequent pressurization with CO2. Using a syringe pump (Isco, 260D), the amount of CO2 required to pressurize the sample to the desired pressure was accurately measured. The loaded sample was

Conclusions

In this study, we describe a magnetically levitated sphere rheometer capable of measuring rheological properties of fluids under supercritical conditions, specifically elevated pressures. This magnetically levitated sphere rheometer eliminates many of the disadvantages associated with other high-pressure rheometers. It can operate at a constant pressure, at all concentration regions from pure polymer to an equilibrated polymer/SCF (supercritical fluid) solution, and at varying shear rates.

Acknowledgements

This material is based upon work supported in part by the STC Program of the National Science Foundation under Agreement No. CHE-9876674. The authors also gratefully acknowledge support from the Kenan Center for the Utilization of Carbon Dioxide in Manufacturing at North Carolina State University and the University of North Carolina at Chapel Hill, and the Office of Naval Research award number N00014-98-1-0157. YJG would also like to acknowledge Elf Atochem North America for financial support.

References (45)

  • L.G Leal
  • J McClain

    Chem Engng

    (2000)
  • L.J Gerhardt et al.

    J Polym Sci Part B Polym Phys

    (1997)
  • C Kwag et al.

    J Polym Sci Part B Polym Phys

    (1999)
  • J.R Royer et al.

    J Polym Sci Part B Polym Phys

    (2000)
  • M Lee et al.

    Polym Engng Sci

    (1999)
  • C.D Han et al.

    J Appl Polym Sci

    (1983)
  • C.D Han et al.

    J Appl Polym Sci

    (1983)
  • C.-Y Ma et al.

    J Cellular Plastics

    (1982)
  • Y.C Bae et al.

    J Appl Polym Sci

    (1997)
  • P.W Bridgman

    Proc Am Acad Arts Sci

    (1926)
  • E Kiran et al.

    J Appl Polym Sci

    (1995)
  • W.D Richards et al.

    Polym Engng Sci

    (1987)
  • F Koran et al.

    J Rheol

    (1999)
  • V.V Semjonow

    Rheol Acta

    (1962)
  • R.F Westover

    SPE Trans

    (1961)
  • M Mackley et al.

    J Rheol

    (1995)
  • M Mackley et al.

    Rheol Acta

    (1996)
  • M.D Hersey et al.

    Mech Engng

    (1928)
  • M Adam

    Revue Phys Appl

    (1984)
  • B Chu et al.

    Rev Sci Instrum

    (1989)
  • R Hilfiker et al.

    Rev Sci Instrum

    (1989)
  • Cited by (0)

    View full text