JBEI team develops one-pot, wash-free process for pretreatment and saccharification of switchgrass; avenues for driving down biofuel cost
Iran-Khodro to introduce cars equipped with its own new diesel engine this year

Sandia team proposes models for partially premixed low-temperature direct-injection diesel combustion

DEC
The Sandia team has proposed extensions to John Dec’s 1997 model for diesel combustion, represented above. Credit: Musculus et al.; Dec. Click to enlarge.

A team at Sandia National Laboratories recently proposed conceptual models for a specific subset of low-temperature combustion regimes: low-load, single-injection, partially premixed compression ignition (PPCI LTC) conditions that are diluted by exhaust-gas recirculation (EGR) to oxygen concentrations in the range of 10–15%.

Their paper, which provides a detailed review and synthesis of various current experimental and modeling studies—and which extends Sandia scientist John Dec’s seminal 1997 model for diesel combustion—is published in the journal Progress in Energy and Combustion Science.

Low temperature combustion refers to a broad range of in-cylinder combustion strategies for the reduction of NOx emissions from diesel combustion; NOx is formed primarily by a thermal mechanism—production rates increase exponentially with temperature. All LTC strategies reduce combustion temperatures by the dilution of the in-cylinder combustible mixtures, either with excess charge gas to create more fuel-lean mixtures, or with moderate to high levels of EGR.

Formation of soot is also slowed by lower combustion temperatures; however, in-cylinder soot oxidation rates slow even faster with the lower temperatures, resulting in a net increase in PM in some conditions. At very high EGR rates, soot formation can be so low that net PM emissions drop; at those extreme EGR rates, however, combustion efficiency is poor, and unburned hydrocarbon (UHC) and CO emissions become excessive.

In their new review and proposal of LTC models, Sandia’s Mark Musculus and colleagues broadly categorized the numerous LTC strategies that have emerged into two groups, according to the degree of pre-mixing.

  • In homogeneous charge compression ignition (HCCI), vaporized fuel is well-mixed with the charge gas prior to compression. HCCI strategies typically employ long in-cylinder mixing times prior to combustion, or event external mixing strategies (e.g., intake port injection), to produce relatively uniform, fuel-lean mixtures, often with minimal use of EGR, especially at low-load. Ignition timing is kinetically controlled, and is decoupled form the timing of the fuel injection event.

    As it has matured, Musculus et al. noted, HCCI has evolved from early uses of very uniform mixtures to intentionally inhomogeneous charge gas mixtures and temperatures for better control of heat release rate. Even for these inhomogeneous HCCI strategies, mixtures are everywhere fuel lean.

  • Partially premixed compression ignition (PPCI) uses direct injection with more moderate mixing times. The charge distributions for PPCI are more heterogeneous at ignition than for HCCI, and include not only fuel-lean but also fuel-rich mixtures. Low combustion temperatures are achieved using EGR. Ignition is more closely coupled to the fuel injection event than with HCCI, though chemical kinetics still play an important role, the researchers noted.

    PPCI LTC strategies can be divided into two subcategories according to fuel-injection and combustion timing, which is either earlier or later than for conventional diesel combustion.

    PPCI strategies have achieved engine-out compliance with NOx and PM limits; however, other emissions (CO and UHC) often exceed regulated limits. Combustion noise can also be an issue, and fuel consumption can be higher than for conventional diesel.

To realize the potential of either early- or late-injection PPCI LTC strategies, the in-cylinder physical and chemical mechanisms leading to the emissions behavior described above must be well understood. For conventional diesel combustion, significant insight into the in-cylinder spray, ignition, combustion, and pollutant-formation processes has been provided by optical diagnostic techniques. Based on observations made using these diagnostics, several conceptual models have been proposed for conventional quasi-steady diesel combustion. In particular, Dec proposed a conceptual model for conventional diesel combustion in 1997. His conceptual model has proven useful as a foundation for understanding and modeling the in-cylinder processes responsible for conventional diesel engine performance and pollutant emissions characteristic.

The objective of this work is to offer extensions of Dec’s conceptual model for typical EGR-diluted low-load PPCI LTC conditions, similarly based on optical diagnostic observations...the experimental observations are complemented by homogeneous reactor simulations using detailed chemical kinetic mechanisms that serve to clarify the impact of finite-rate chemistry on the ignition and combustion process for PPCI LTC conditions.

—Musculus et al.

The proposed conceptual models are not intended to describe all LTC strategies, Musculus and his colleagues emphasized, but a common subset with the following characteristics:

  1. LTC is achieved by EGR, with intake oxygen concentrations in the range of 10-15%. Under such dilution levels, combustion temperatures are still high enough that soot may form in fuel-rich mixtures.

  2. Fuel injection timing may be either early- or late-injection PPCI. In either case, the ignition dwell is positive so that fuel is partially premixed before combustion. The ignition dwell is short enough, however, that ignition is still somewhat coupled to the fuel injection timing, and the residual jet structure is often still intact at ignition.

  3. The fuel injector uses conventional small-orifice diesel nozzles to produce typical diesel-like sprays.

  4. Fuel is delivered in a single-injection event—i.e., no pilot, split or post injections.

  5. Fuels are diesel-type, with two-stage ignition chemistry—i.e., not gasoline-like fuels.

Measuring CO and UHC
The Sandia research team identified the sources of carbon monoxide (CO) and unburned hydrocarbons (UHC) from LTC engines using new optical diagnostic techniques.
The researchers used two-photon laser-induced fluorescence to map in-cylinder CO, a difficult measurement that had never been achieved inside a diesel engine.
Detecting UHC is problematic because UHC comprises many different chemical species, with composition evolving during combustion. Instead of detecting UHC directly, researchers used laser-induced fluorescence of other markers of combustion, such as formaldehyde and hydroxyl, to observe and understand the chemical processes that lead to UHC.
The combined measurements showed that the fuel that ended up near the fuel injector was “over-mixed”—there was too much air and not enough fuel, leading to the CO and UHC in the exhaust.
The Sandia team has suggested post-injections to increase fuel concentration in that area. With the post-injections, the zone of complete combustion extends over a larger region, leading to lower UHC and CO emissions while increasing efficiency.

Based on observations from optical diagnostics, chemical kinetics simulations, and reviews of earlier studies, the Sandia team offered a set of models extending the Dec model under highly dilute (~13% intake O2), low-load (~4 bar gIMEP) PPCI conditions with ignition occurring at or after the end of injection.

The first model is applicable to heavy-duty engines, in which the fuel-jet structure is less perturbed in in-cylinder surfaces and flows. The second pair is for light-duty engines with either early or late injection, in which in-cylinder surfaces and flows do significantly deflect the fuel jets.

The models describe the spray formation, vaporization, mixing, ignition, and pollutant formation and destruction mechanisms.

Relative to the existing conventional diesel conceptual model, the features of the LTC conceptual models include longer liquid-fuel penetration; an extended ignition delay that allows more premixing of fuel; a more distinct and temporally extended two-stage ignition; more spatially uniform second-stage ignition; reduced and altered soot formation regions; and increased overmixing leading to incomplete combustion.

Musculus
Conceptual model for single early-injection, low-load, EGR-dilute, PPCI low-temperature light-duty DI diesel combustion. Musculus et al. Click to enlarge.

The Sandia work was completed for the US Department of Energy’s Office of Energy Efficiency and Renewable Energy (EERE).

Mark Musculus discusses the work at Sandia’s Combustion Research Facility on Low Temperature Combustion regimes.

Resources

  • Mark P.B. Musculus, Paul C. Miles, Lyle M. Pickett (2013) Conceptual models for partially premixed low-temperature diesel combustion, Progress in Energy and Combustion Science, Volume 39, Issues 2–3 Pages 246-283 doi: 10.1016/j.pecs.2012.09.001

  • Dec, J. (1997) A Conceptual Model of DI Diesel Combustion Based on Laser-Sheet Imaging, SAE Technical Paper 970873 doi: 10.4271/970873

Comments

Richard Batty

Good research but it cost a lot of money. Better if we switch to DME as a fuel and do the research on improving the combustion of DME. On the other hand, under normal engine conditions, DME burns completely, produces no soot and will produce very little NOx with a little tuning.

The comments to this entry are closed.