Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Marangoni”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 records
All sourcesNASA NTRSDOE OSTI
NASA NTRS1985

Suppression of Marangoni Convection in Float Zones

The basic purpose of this program is to demonstrate by means of an Earth-based 1-g experiment that the undesirable Marangoni (surface tension) convection can be suppressed or significantly reduced by means of gas jets directed tangentially to the free surface of the liquid in a float zone. These jets will establish the tangential shear stress field over the surface which must be adjusted to equal the counter-stress resultant of the Marangoni shear stress which causes the convection. For proposed materials processing in space (o-g), particularly of important, highly reactive semiconductor materials, e.g., silicon, microgravity will virtually eliminate the unwanted thermal-buoyancy convection in the liquid silicon, but will have no effect in reducing the Marangoni convection. Unless this can be sufficiently suppressed by other means, there may be no significant advantages to the proposed space processing of reactive semiconductors. Although some inert gas such as argon must be used for the corrosive liquid silicon, the Earth-based experiment uses air jets and various transparent oils, since the basic principle involved is the same. The first float zone is enclosed in a very small rectangular box with a quasi-planar free surface. Stable Marangoni convection has been achieved and velocities measured photographically. The air jet system with variable velocity and temperature is under construction. Three independent parameters must be optimized to attain maximum suppression: the gas velocity, angle of attack, and gas temperature.

Robert F Dressler↗
NASA NTRS1979

Marangoni bubble motion in zero gravity

It was shown experimentally that the Marangoni phenomenon is a primary mechanism for the movement of a gas bubble in a nonisothermal liquid in a low gravity environment. A mathematical model consisting of the Navier-Stokes and thermal energy equations, together with the appropriate boundary conditions for both media, is presented. Parameter perturbation theory is used to solve this boundary value problem; the expansion parameter is the Marangoni number. The zeroth, first, and second order approximations for the velocity, temperature and pressure distributions in the liquid and in the bubble, and the deformation and terminal velocity of the bubble are determined. Experimental zero gravity data for a nitrogen bubble in ethylene glycol, ethanol, and silicone oil subjected to a linear temperature gradient were obtained using the NASA Lewis zero gravity drop tower. Comparison of the zeroth order analytical results for the bubble terminal velocity showed good agreement with the experimental measurements. The first and second order solutions for the bubble deformation and bubble terminal velocity are valid for liquids having Prandtl numbers on the order of one, but there is a lack of appropriate data to test the theory fully.

Thompson, R. L.↗
NASA NTRS1981

Eigenvalues of the Rayleigh-Benard and Marangoni problems

The eigenvalues of the linear Benard-Marangoni stability problem are discussed. Pearson and Nield boundary conditions, which correspond to a rigid, isothermal lower boundary and a stress-free conducting upper boundary are considered. It is shown that although a critical value of the Marangoni number can be determined, the number is not, strictly speaking, an eigenvalue and cannot be used as an eigenvalue parameter for the determination of an eigenvector set.

Rosenblat, S.↗
NASA NTRS1982

Nonlinear Marangoni convection in bounded layers. I - Circular cylindrical containers. II - Rectangular cylindrical containers

Liquid undergoing nonlinear Marangoni instability in a circular cylinder is examined, with attention given to roll-cell development and interaction. Surface deflections are neglected and the side walls are considered as adiabatic and impenetrable, allowing the liquid to freely slip. The nonlinear convective states are calculated and their stability is defined. The behavior and amplitude of cells forming in the liquid, heated from below, are modeled in order to derive all the transport properties. A new small parameter is formulated which is related to the critical Marangoni number of the infinite matrix expressing the eigenvalue expansion of the problem. The observed roll cell amplitudes and transport properties are shown to be available from simple eigenvalues, with double eigenvalues, indicating the existence of two roll-states as predicted by linear theory, nonlinear theory indicates transitions from one steady convective state to another.

Rosenblat, S.↗
NASA NTRS1974

Ultrapure materials - Containerless evaporation and the roles of diffusion and Marangoni convection

Contamination from containers is a major problem in preparing ultrapure refractory materials. Space with its zero gravity and its high vacuum offers an opportunity for containerless purification of these materials. The evaporation of impurities from a melt will involve many complex chemical equilibria. Thermodynamic calculations have been modified to describe these equilibria when impurities in the melt evaporate into vacuum. The contributions of diffusion and Marangoni convection to mass transfer rates in the bulk liquid have been estimated. Calculations for the evaporative purification of molten alumina are given.

Dragoo, A. L.↗
NASA NTRS1979

Marangoni bubble motion in zero gravity

It is shown experimentally that the Marangoni phenomenon is a primary mechanism for the movement of a gas bubble in a nonisothermal liquid in a low-gravity environment. In such two-phase flow systems, local variations in bubble surface tension are caused by a temperature gradient in the liquid. Shearing stresses thus generated at the bubble surface lead to convection in both media, as a result of which the bubble begins to move. A mathematical model consisting of the Navier-Stokes equations and the thermal energy equations, along with the appropriate boundary conditions for both media, is proposed.

Thompson, R. L.↗
NASA NTRS1981

The Marangoni effect and translation of free non-deformable drops

A model is presented for flow caused by interface tension gradients, the so called Marangoni effect, on a free, nondeformable drop. A free drop, initially at rest, undergoes a translation motion upon the action of surface flow. The experiments carried out by injecting a drop with surfactants, which induce an interface tension gradient, are in good agreement with the theoretical model proposed.

Chifu, E.↗
NASA NTRS1974

The effect of temperature induced surface tension gradients on bubble mechanics

Experiments were conducted to examine in detail the influence of surface tension gradient induced flow, the Marangoni effect. One set of studies provided visual results that demonstrated the occurrence and magnitude of the Marangoni effect. Another portion of the work was directed to bubble force measurements. These studies used in deflection of a very fine cantilevered wire with an attached bubble as a means of measuring force with and without temperature gradients. The experimental force data were found to check existing theoretical predictions reasonably well except for the case where the test fluid had a sizeable vapor pressure. In this situation the thermophoretic force on the bubble was much higher than predicted.

Mcgrew, J. L.↗
NASA NTRS1980

Influence of solidification on surface tension driven convection

The effect of solidification on the onset of surface tension driven convection in a reduced gravity environment is studied. Two simple but physically realistic configurations representing the solidification of a simple material are analyzed. The analysis shows that as a result of the solidification process the critical Marangoni number is shifted to lower values indicating that solidification has a destabilizing effect upon the liquid. From this result it is concluded that convection can be brought about in the liquid phase at lower Marangoni numbers when solidification is present than when it is not. The effects of other parameters introduced by the solidification process are analyzed and discussed.

Antar, B. N.↗
NASA NTRS1977

CTS TEP thermal anomalies: Heat pipe system performance

A part of the investigation is summarized of the thermal anomalies of the transmitter experiment package (TEP) on the Communications Technology Satellite (CTS) which were observed on four occasions in 1977. Specifically, the possible failure modes of the variable conductance heat pipe system (VCHPS) used for principal thermal control of the high-power traveling wave tube in the TEP are considered. Further, the investigation examines how those malfunctions may have given rise to the TEP thermal anomalies. Using CTS flight data information, ground test results, analysis conclusions, and other relevant information, the investigation concentrated on artery depriming as the most likely VCHPS failure mode. Included in the study as possible depriming mechanisms were freezing of the working fluid, Marangoni flow, and gas evolution within the arteries. The report concludes that while depriming of the heat pipe arteries is consistent with the bulk of the observed data, the factors which cause the arteries to deprime have yet to be identified.

Marcus, B. D.↗
NASA NTRS1978

Convection due to surface-tension gradients

The use of dimensionless parameters to study fluid motions that could occur in a reduced-gravity environment is discussed. The significance of the Marangoni instability is considered, and the use of dimensionless parameters to investigate problems such as thermo and diffusocapillary flows is described. Characteristics of fluid flow in space are described, and the relation and interaction of motions due to capillarity and buoyancy is examined.

Ostrach, S.↗
NASA NTRS1978

Solidification (crystal growth) in the presence of gravitational forces

The surface tension behavior of doped and undoped InSb melts was investigated as well as their temperature and composition dependence. Surface tension in InSb melts was determined using the sessile-drop technique covering the temperature range from 530 C to 880 C. A linear regression of the data obtained shows that the temperature dependence of sigma is 392- (T-530) x (7000) plus or minus 10 dyne/cm. The d sigma/d Tau for intrinsics InSb is less than that previously reported. On the basis of the surface tension data obtained, it is concluded that surface tension induced convective flow velocities in InSb under reduced gravity conditions range from zero to at most 1 cm/sec. Accordingly, no interference with dopant segregation can be expected during growth in space because the momentum boundary layer (at the crystal melt interface) associated with any Marangoni-type convective flows would, at the given growth rate, be significantly larger than the predicted diffusion boundary layer thickness.

Gatos, H. C.↗
NASA NTRS1981

The nature of melt process changes in microgravity

The effects on melt impurities of the lack of buoyancy-driven convection in melts and the gas phase are outlined. Probable effects noted are: retardation of convective flow by a thin adsorbed impurity layer because of surface tension gradients (Marangoni flow); changes in the transport of volatile impurities to the melt surface and the consequent reduction of evaporation; and the change in dependence of evaporation rates on pressure due to the absence of convective flow in the gas next to the melt.

Wilcox, W.↗
NASA NTRS1981

Review of literature surface tension data for molten silicon

Measurements of the surface tension of molten silicon are reported. For marangoni flow, the important parameter is the variation of surface tension with temperature, not the absolute value of the surface tension. It is not possible to calculate temperature coefficients using surface tension measurements from different experiments because the systematic errors are usually larger than the changes in surface tension because of temperature variations. The lack of good surface tension data for liquid silicon is probably due to its extreme chemical reactivity. A material which resists attack by molten silicon is not found. It is suggested that all of the sessile drip surface tension measurements are probably for silicon which is contaminated by the substrate materials.

Hardy, S.↗
NASA NTRS1982

Use of thermocapillary migration in a controllable heat valve

In accordance with the Marangoni effect, immiscible droplets in a host fluid in which a temperature gradient exists move in the direction of increasing temperature. It is proposed that this thermocapillary migration could be used to construct a 'liquid wick' that would return the condensed vapor at the condenser end of a heat pipe back to the evaporator, thus completing the fluid circuit. The droplets would be formed by capillary pressure forcing the condensate through a perforated diaphragm whose temperature would control the droplet flux, and hence the heat flux between the two ends of the heat pipe, thus making it a controllable heat valve.

Schmid, L. A.↗
NASA NTRS1995

The Benard Problem: A Comparison of Finite Difference and Spectral Collocation Eigen Value Solutions

The application of spectral methods, using a Chebyshev collocation scheme, to solve hydrodynamic stability problems is demonstrated on the Benard problem. Implementation of the Chebyshev collocation formulation is described. The performance of the spectral scheme is compared with that of a 2 nd order finite difference scheme. An exact solution to the Marangoni-Benard problem is used to evaluate the performance of both schemes. The error of the spectral scheme is at least seven orders of magnitude smaller than finite difference error for a grid resolution of N = 15 (number of points used). The performance of the spectral formulation far exceeded the performance of the finite difference formulation for this problem. The spectral scheme required only slightly more effort to set up than the 2 nd order finite difference scheme. This suggests that the spectral scheme may actually be faster to implement than higher order finite difference schemes.

J Raymond Lee Skarda↗