Begell House Inc.
Atomization and Sprays
AAS
1044-5110
10
3-5
2000
Letter from the Editor
iii
10.1615/AtomizSpr.v10.i3-5.10
Ten years ago, it was suggested that ILASS have a journal for spray research and applications. Many of us had experienced difficulty persuading journal editors to accept papers on sprays. An occasional paper found its way to publication in widely disparate journals that were difficult to access. Having a journal of our own would not only give us access to publication, but would also provide a focus to encourage the writing of papers on sprays. We approached several publishers who told us that in the climate of reduced budgets for libraries and increased costs of publication, journals were being closed down and there was no room for a new journal on sprays.
Bill Begell had just started a small publishing company and he was willing to take a gamble, based on his belief that spray technology would grow—much like the growth of two-phase flow. We started publication with four issues per year, and as the flow of submitted manuscripts increased, the number of issues per year climbed to six. Persuading institutional libraries and individuals to subscribe to our journal had limited success until ILASS-Americas decided to include subscription to the journal in the registration fee at their annual conferences. This provided a big boost to subscriptions and has saved the journal from extinction. ILASS Asia and Europe should be strongly encouraged to include journal subscription in their own conference registration fees.
The success of our journal can be measured by the continuous flow of manuscripts submitted, both from within the regional ILASS communities and others seeking a venue for publication. Also, the high level of citations is important for our academic colleagues seeking promotion. Our papers and readership cover a wide spectrum: physics, fluid and particle mechanics, computational fluid dynamics, mechanical and chemical engineering, fuel injection, spray drying, coatings, food processing, pharmaceutical coatings of tablets, aerosols for inhalation therapy, etc. Whereas fuel injection is at the most advanced level of theoretical and experimental analysis, other disciplines and technologies are beginning to use laser diffraction, imaging, and phase Doppler instrumentation for characterizing sprays and determining size and velocity distributions. As more industries hire engineers with computer skills who can use advanced codes for turbulent two-phase flows, such as Fluent and KIVA, industries are learning that computational analysis can provide detailed insight and predictions of fluid flow and particle characteristics.
CD roms and diskettes are leading the way toward electronic publishing, and journal articles are now available on the internet. ICLASS regional, and national ILASS and their annual conferences, combined with regular publication of the journal, have established Atomization and Sprays as a significant milestone in the field of spray science publication.
Letter from the Publisher
iv
10.1615/AtomizSpr.v10.i3-5.20
The history of science and engineering shows that progress is frequently, if not primarily, achieved through the means of serendipitous discovery. Atomization and Sprays, as both a decade-old archival journal and a branch of engineering science, was essentially conceived through a serendipitous acquaintance between its founding editor and the publisher several years before its launch, ultimately leading to the beginning of the publication.
There are many examples of now well-established areas of science and engineering that were originally splintered from basic and fundamental disciplines. This is a natural phenomenon of growth and development.
In scholarly publishing circles, such growth and development has been known for about forty years as "twigging," the term having been coined by Curtis Benjamin, the late president of McGraw-Hill. As the term implies, large trunks of knowledge develop branches that, in turn, sprout twigs. The entire tree continues to grow, with the older trunks frequently crumbling away, the newer branches growing stronger and thicker, and the twigs becoming branches and giving birth to new little twigs. As with all things in nature, some twigs survive and others do not. (Do we still remember the name of the bulbous contraptions in our radios before the advent of the semiconductor?)
Another natural occurrence in engineering and science is the development of inter-disciplinary and cross-disciplinary specialties, such as analytical physical chemistry and biomedical engineering, respectively.
Some of them grow, develop, and thrive; others—due to economic, marketing, or scientific reasons—die on the vine.
From the publisher's point of view, the launching of a new journal, in a twigging stage of growth and both interdisciplinary and cross-disciplinary orientations, takes commercial guts. The most difficult and most expensive part of such an enterprise is the marketing and targeting of the audience. In the case of Atomization and Sprays, how does one fish out the engineer or pharmacologist who works for a pharmaceutical company manufacturing antiwheezing sprays? You need to spend oodles of dollars to find one possible subscriber among the thousands from available lists and directories. Tough decision, tough going. Somehow, my intuition was correct.
And we made it. With the help of ICLASS, ILASS, our Editorial Boards, and last but not least, Professor Norman Chigier, the indefatigable atomization-and-sprays man, we have reached the right audience and have developed a primary, important, frequently cited, archival engineering journal: Atomization and Sprays. Serendipitously.
SPRAY AND COMBUSTION MODELING IN GASOLINE DIRECT-INJECTION ENGINES
219-249
10.1615/AtomizSpr.v10.i3-5.30
Li
Fan
Engine Research Center, University of Wisconsin, Madison, Wisconsin, USA
Rolf D.
Reitz
Engine Research Center, University of Wisconsin-Madison, Rm 1018A, 1500 Engineering Drive, Madison, Wisconsin 53706, USA
Computer simulation models for fuel preparation and combustion in gasoline direct-injection spark-ignition (GDI or DISI) engines are described. A modified KIVA-3V code that includes improved spray breakup, wall impingement, and combustion models was used. In particular, a new ignition kernel model, called DPIK (discrete particle ignition kernel), has been developed to describe the early flame kernel growth process. The model uses Lagrangian marker particles to describe the flame kernel location. The spray and engine flow models were validated using available drop size and patternator measurements, and particle tracking velocimetry (PTV) data from a water analog rig. The combustion models were applied and validated for both homogeneous and stratified-charge engines. The stratified-charge engines considered include both wall-guided and spray-guided designs. Applications of the models show that optimized injection timing can lead to reduced wall wetting, higher turbulence intensity near TDC, and better volumetric efficiency and knock resistance. For wall-guided combustion chamber designs, the injector orientation significantly influences the fuel stratification pattern, and hence the combustion characteristics. The gas tumble also affects the fuel distribution and the ignition process. Under certain conditions, the fuel—air mixing is characterized by the existence of many lean regions in the cylinder and the burning speed is very low; hence the combustion can be poor in these cases. Multidimensional modeling is shown to be a useful tool to help visualize and optimize engine combustion details.
FIFTY YEARS OF GAS TURBINE FUEL INJECTION
251-276
10.1615/AtomizSpr.v10.i3-5.40
Arthur H.
Lefebvre
Emeritus Professor, Cranfield University, Stratford, U.K., and Purdue University, W. Lafayette, IN, USA
As its title suggests, this article is devoted to developments in gas turbine fuel injection during the past half-century. It describes in general terms the evolution of pressure atomizers from the simplex nozzle of the 1940s to the dual-orifice injector that remained in widespread use for over 20 years until it was replaced by the various forms of airblast atomizer that dominated the scene for the next three decades. The pressure nozzles described herein include simplex, duplex, dual-orifice, fan-spray, and spill-return. The inherent design flexibility of the airblast concept encouraged a wide variety of injector configurations, ranging from simple air-assist nozzles to the more sophisticated designs of today, in which part of the atomizing air is carried by the nozzle itself while the remainder flows through swirlers mounted on the combustion liner. Attention is focused on the relative merits of the various nozzle types, both pressure and airblast, in regard to their ability to satisfy stringent performance requirements while surviving for many thousands of hours in the increasingly hostile environment created by the continuing trend toward engines of higher pressure ratio. Reference is made to some new developments in atomizer design and manufacture and to the ongoing role of the fuel injector in finding solutions to the problems posed by ultralow-emissions combustors, many of which are required to operate near the lean extinction limit on fully premixed fuel−air mixtures.
GENERALIZED LORENZ-MIE THEORIES, FROM PAST TO FUTURE
277-333
10.1615/AtomizSpr.v10.i3-5.50
G.
Gouesbet
Laboratoire d'Energetique des Systemes et Precedes (LESP), UMR CNRS 6614, CORIA, INSA de Rouen, and University of Rouen, Mont-Saint-Aignan Cedex, France
Gerard
Grehan
CORIA, Universite de Rouen, Site Universitaire du Madrillet BP 12 76801 Saint Etienne du Rouvray, France
We review the generalized Lorenz-Mie theories developed during the two last decades, describing the interaction between arbitrary-shaped beams (generic laser beams) and some regular particles, as well as the applications which have been implemented. We take this opportunity to attempt a forecasting of further studies that should be also developed in the same line of research.
TOWARD A COMPREHENSIVE THEORY OF DENSE SPRAY FLOWS
335-353
10.1615/AtomizSpr.v10.i3-5.60
Christopher F.
Edwards
Thermosciences Division, Department of Mechanical Engineering, Stanford University, Stanford, CA, USA
A mathematical theory of dense spray flows is described. It is based on treating the flow both inside and outside the drops via the incompressible Navier-Stokes equations and the interface as Gibbs' dividing surface. Taking into account the information necessary to describe the dynamical state of the flow, a hyperspace is constructed which describes the state of the system at any instant. This hyperspace consists of a number of field axes—one which describes the instantaneous velocity field and a number which describe the instantaneous morphology of the fluids.
Following the methods of statistical mechanics, an ensemble of macroscopically identical flows is used to define a density of system points in the hyperspace. A transport equation is then written which describes the evolution of this collection of flows for all time. A unique feature of this transport equation is that the dynamics of each fluid element are embedded in the transport equation—that is, the Navier-Stokes equations and interface jump conditions are implicit constraints on the overall transport of a system point in hyperspace.
The utility of the resulting equation (the continuum-particle, continuum-field equation) is demonstrated by showing that it can be reduced, in the limit of small dispersed-phase elements, to the point-particle, continuum-field equation which, in turn, has been shown to reduce to the ensemble-averaged Navier-Stokes and spray equations. As such, the present development is the uppermost level of a hierarchy of models for continuum treatment of spray flows—analogous to the Liouville equation of the kinetic theory of gases.
VIEWS ON THE STRUCTURE OF TRANSIENT DIESEL SPRAYS
355-386
10.1615/AtomizSpr.v10.i3-5.70
Gregory J.
Smallwood
Aerosol and Gas Metrology, Metrology Research Centre, National Research Council, Ottawa, Ontario, Canada K1A 0R6
Omer L.
Gulder
National Research Council Canada, ICPET, Combustion Research Group, Ottawa, Ontario, Canada
There has been tremendous change over the last few decades in the operating conditions of diesel fuel injection systems and engines, and in the diagnostic tools and numerical models available to evaluate them. Improvements in the diagnostic techniques coinciding with changes in diesel injector technology have brought about an entirety different view of the breakup of liquid in current diesel sprays. A detailed examination of the history and current understanding of the structure of the dense core region in transient diesel sprays is presented.
Diagnostic methods are reviewed, and the appropriate uses are discussed. Of the techniques currently available, tomography is the most appropriate for determining the structure of the dense core region. Conductivity is not recommended. Line-of-sight techniques are recommended only for studying the periphery of the spray. Due to its greater contrast, high-intensity Mie scattering is preferred over line-of-sight methods for liquid spray penetration distance measurements. Advances in phase-Doppler interferometry are required to provide drop size and velocity measurements in the near-nozzle region.
A review of the spray structure and breakup mechanisms is presented. The structure of the spray has been shown to be completely atomized at or near the nozzle tip, with nozzle cavitation and turbulence instabilities as the dominant breakup mechanisms. Buckling may be responsible for breakup during the very early phase of injection. Aerodynamic shear may cause some secondary atomization, but its role in breakup is far less significant than previously thought. Cavitation affects jet breakup through the bursting and collapsing vapor cavities, thus contributing to the disintegration of liquid, resulting in a mixture of bubbles and liquid occupying most of the cross-sectional area, and through increasing the turbulence intensity, thus contributing to the instability of the liquid jet. The turbulence instability, along with pressure fluctuations in the nozzle, cause variation in the exit velocity of the droplets, resulting in temporal and spatial clustering of the droplets in the plume.
The results of recent research on the liquid spray penetration distance and drop size have been summarized. For liquid spray penetration distance, the orifice diameter is the dominant injection parameter, and ambient density is the dominant engine parameter, although ambient temperature is also significant. Fuel properties have been shown to have an effect on the liquid spray penetration distance, but further research is required to draw significant conclusions. For drop size, injection pressure and orifice diameter are the known dominant parameters.
The evidence for complete atomization of diesel sprays near the nozzle has come from a variety of sources, including tomographic imaging of the internal structure, microphotography of the near-nozzle region, diffraction droplet sizes that are greater on the periphery than the centerline, infrared multiwavelength extinction droplet sizing, and internal flow studies.
MODELING OF SPRAY IMPACT ON SOLID SURFACES
387-408
10.1615/AtomizSpr.v10.i3-5.80
Cameron
Tropea
Technische Universitat Darmstadt, Institute for Fluid Mechanics and Aerodynamics, Alarich-Weiss-Str. 10, 64287 Darmstadt, Germany
Ilia V.
Roisman
Technische Universität Darmstadt, Institute of Fluid Mechanics and Aerodynamics, Center of Smart Interfaces, Germany
This work presented in this article differs from conventional approaches in modeling spray impact on walls and films by departing from the idea that results obtained from single drop impacts can be simply extrapolated to the case of sprays. An empirical model of spray impact on solid surfaces accounting for the interaction of neighboring impacts is proposed. Propagation of a crown resulting from the impact of a single drop is analyzed theoretically, and a statistical parameter λ, characterizing the occurrence probability of crown interactions on the surface, is estimated. Then, the model for single drop impacts is corrected using the parameter λ to fit the experimental results of spray impact. A simple form of the probability density function of the secondary droplets is proposed, and analytical expressions for its parameters are given. It is also shown experimentally that the behavior of the spray at a given point near the solid surface can be influenced by conditions far from this point.
PERSPECTIVES ON LARGE EDDY SIMULATIONS FOR SPRAYS: ISSUES AND SOLUTIONS
409-425
10.1615/AtomizSpr.v10.i3-5.90
Josette
Bellan
Department of Mechanical and Civil Engineering, Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California, USA
A review of the large eddy simulations (LES) methodology is presented in the context of sprays. Issues related to modeling both the drop interaction with the carrier flow and the interaction among drops are discussed. Appropriate direct numerical simulations (DNS) for use as precursors to LES, and the extraction of subgrid scale (SGS) models are both described. Particular attention is devoted to LES aspects which are different from those of single-phase flows. These include the correct portrayal of the drop interaction with small turbulent scales, the modeling of SGS stresses, SGS heat and SGS species fluxes, and the accurate representation in the carrier flow equations of the source terms associated with the presence of the drops. Recommendations for future work are also offered.
FUEL ATOMIZATION FOR NEXT-GENERATION GAS TURBINE COMBUSTORS
427-438
10.1615/AtomizSpr.v10.i3-5.100
The push toward higher specific fuel consumption and smaller, lighter packaging for reduced-cost aerospace gas turbine engines has resulted in large increases in engine operating pressures and temperatures, as well as major efforts to reduce gas path losses and increase component efficiencies. This is a trend that is expected to continue, and as a result, thermal management of the hot engine section, including the fuel nozzle, combustor, and turbine, has emerged as a critical technology area requiring further research and development. For the fuel injection system, nozzle thermal management, turndown ratio, and atomization performance while maintaining correct combustor aerodynamics and low pollutant emissions are the most important performance features that necessitate optimization. Complex and expensive heat-shielded designs are often required to reduce nozzle wetted-wall temperatures and prevent the formation of carbonaceous deposits within the fuel delivery passages. Optimization of designs using current computational methods is limited in capability, and expensive. Significant advances in fuel injection concepts, physical understanding, and computational methods are required to meet these increasingly demanding combustor requirements, with configurations at or below current cost levels. Five injector designs are presented, which include an advanced hybrid air blast (HAB) atomizer, a lean direct-injection (LDI) concept, and three lean prevaporized premixer (LPP) designs that exemplify advanced fuel injection technology and ideas to address the challenges of next-generation gas turbine combustors.
SPRAY DIAGNOSTICS FOR THE TWENTY-FIRST CENTURY
439-474
10.1615/AtomizSpr.v10.i3-5.110
William D.
Bachalo
Artium Technologies, Inc., Aerometrics, Inc. Sunnyvale, 150 West Iowa Avenue, Unit 202, Sunnyvale, California, USA
A critical review of the progress in atomization and spray technology is presented. Although significant progress has been made with the benefit of developments in theory, diagnostics, and modeling, the ability to accurately prescribe or predict the spray and related two-phase turbulent flow behavior in detail has generally eluded our efforts. Available diagnostics are reviewed and some extrapolations are offered suggesting where evolving technologies in the areas of electronics, computers, software, and information technology (IT) might lead. Experimentation conducted over the past decades is criticized for lacking completeness and quality in the information measured and for its failure to identify, measure, and record all of the significant parameters. Deficiencies in measurement capabilities are recognized and arguments are presented regarding the need to integrate the experimentation and modeling. Proposed strategies include the simultaneous experimentation with interactive model predictions, which may be the only means available to completely describe the spray processes and attain the development needed in actually predicting the phenomena. Currently, the typical time scale for interactions between experiments and numerical modeling are of the order of years. Elapsed time needs to be reduced to seconds and, in many cases, to real time. Reaching the goal of accurate prediction, understanding, and the ability to prescribe spray characteristics will require highly automated data acquisition, integrated modeling, storage, and efficient, easy access to large volumes of information. This will require even greater reliance on computing power, information systems, and the Next Generation Internet (NGI).
IMPROVING DROPLET BREAKUP AND VAPORIZATION MODELS BY INCLUDING HIGH PRESSURE AND TURBULENCE EFFECTS
475-510
10.1615/AtomizSpr.v10.i3-5.120
Iskendar
Gökalp
CNRS-INSIS, Institut de Combustion, Aérothermique, Réactivité et
Environnement, Orléans, 45071, France;
Department of Mechanical Engineering, Middle East Technical University,
Ankara, 06800, Turkey
Christian
Chauveau
CNRS-INSIS, Institut de Combustion, Aérothermique, Réactivité et
Environnement, Orléans, 45071, France
Celine
Morin
UVHC, TEMPO, 59313 Valenciennes, France
B.
Vieille
Centre National de la Recherche Scientifique, Laboratoire de Combustion et Systemes Reactifs, 45071 Orleans Cedex 2, France
Madjid
Birouk
Department of Mechanical and Manufacturing Engineering, University of Manitoba, Winnipeg, MB, R3T 5V6 Canada
This article reviews recent experimental work conducted at the Laboratoire de Combustion et Systemes Reactifs (LCSR), Orleans, France, on single droplet breakup and vaporization. Emphasis is essentially put on high pressure and turbulence effects. The experimental facilities developed and the diagnostics used are first presented. Droplet breakup studies are conducted with cryogenic and noncryogenic droplets subjected to aerodynamic shear forces under high-pressure conditions. The transition criteria between droplet breakup regimes, characteristic breakup times, and secondary droplet distributions are obtained for uniquely low values of the density ratio between liquid and gas phases and systematically varied values of droplet Weber and Reynolds numbers. Combined effects of high pressure and temperature on droplet vaporization are also systematically explored. The variation patterns of average vaporization rates with reduced pressure and temperature are conclusively established and compared to estimates from the quasi-steady model. The influence of turbulence on droplet vaporization rates is explored in detail. It is demonstrated that droplet vaporization rates increase significantly with turbulent Reynolds number, even when the droplet size is smaller than the turbulence integral length scale. Comprehensive correlations are established to take into account these various effects. Suggestions are made for ways of including these correlations as submodels into spray combustion numerical prediction codes and for future work to further improve them.
SPRAY BREAKUP MECHANISM FROM THE HOLE-TYPE NOZZLE AND ITS APPLICATIONS
511-527
10.1615/AtomizSpr.v10.i3-5.130
Hiroyuki
Hiroyasu
Institute of Industrial Technology, Kinki University, Higashi-Hiroshima, Japan
The fundamental physical processes of the spray breakup mechanism from the hole-type nozzle are examined. In early research it was believed that atomization of the liquid jet was caused by the interfacial forces existing between the issuing jet and the surrounding gas. However, investigation has determined that the strong turbulence in the nozzle hole due to cavitation phenomena contributes greatly to the disintegration of the liquid jet. To reveal the mutual relationships, experiments were performed under conditions with varying length-to-hole diameter ratios L/D, and different inlet shapes and different internal shapes of the nozzle. As a consequence of this study, it has been determined that the primary factor in atomization of the liquid jet is the disturbance of the liquid flow resulting from cavitation phenomena. Next, the effects of the internal flow in a diesel nozzle on the atomization of a spray were analyzed experimentally and numerically. Flow visualization studies were made using a transparent acrylic model nozzle.