Date of Award

Spring 5-15-2018

Author's School

School of Engineering & Applied Science

Author's Department

Energy, Environmental & Chemical Engineering

Degree Name

Doctor of Philosophy (PhD)

Degree Type



Innovative coal technologies are essential for addressing concerns about air pollution and global climate change. A key pathway to advancing these technologies is through developing a thorough understanding of the fundamental physical and chemical processes that occur during coal combustion. Ignition influences many aspects of coal combustion, including flame stability, submicron aerosol evolution, and char burnout. As important as ignition and these associated processes are, they are challenging to study because they depend on many factors, such as the combustion environment, particle size, and particle-particle interactions.

While there have been many studies of coal ignition, none have studied the process in a way that simulates the coal particles going from a reducing to an oxidizing environment, which is characteristic of what coal particles experience in the near-burner region of pulverized coal furnaces. Fundamental studies of the impacts of this “reducing-to-oxidizing” environment on ignition as a function of residence time, gas temperature, and particle size can provide valuable insights for optimizing advanced burner design.

In addition to ignition, ultrafine aerosols are mainly formed in the near-burner region of pulverized coal furnaces where temperature is high, and a better understanding of the formation mechanisms in the reducing-to-oxidizing environment can aid the development of in-flame control of particulate formation. Thus, in this dissertation, a new experimental platform, called a two-stage Hencken flat-flame reactor, is designed and fabricated for evaluating coal ignition and aerosol formation in conditions relevant to pulverized coal furnaces with respect to timescales, gas temperatures, and combustion environments.

With this new reactor, particles first experience a reducing environment and then transition to an oxidizing environment, in a residence time characteristic of typical coal furnaces. Using this unique experimental apparatus, single particle ignition and coal stream ignition data are obtained under a wide range of conditions. The methods to obtain these data include high-speed videography, image-processing technique (particle tracking and intensity extraction), and scanning electron microscopy. Together with computational model predictions, characteristics of single particle ignition and coal stream ignition are obtained which include ignition mechanisms, ignition delay times and induction times.

First, a proof-of-concept study is performed to evaluate the influence of the “reducing-to-oxidizing” environment (R-O) on single particle ignition. The proof-of-concept study is repeated for single particles experiencing an oxidizing environment only (O). The results show that the R-O affects single particle ignition significantly. Specifically, at a high gas temperature of 1800 K, a hetero-homogeneous mechanism is promoted in the R-O while a homogeneous-to-heterogeneous mechanism prevails in the oxidizing environment. For 1300 K gas temperature environment, it is found that volatiles are released mainly after the particle has transitioned to the oxidizing environment, thus promoting homogeneous ignition. Due to the R-O, average ignition delay times for single particles in nominal 1300 K and 1800 K gas temperatures increase over those of O by 20% and 40% respectively.

Next, the role of different particle sizes on single particle ignition is studied. Unique to the R-O, ignition delay times for particles above 106 µm size are found to be similar in 1800 K gas temperature. The similarity is due to high volatile fluxes from such large particles. Hence, homogeneous ignition occurs as soon as such a particle with its volatile transitions to the oxidizing environment from the reducing environment. In a 1300 K gas temperature environment (lower heating rate), the ignition delay times in O and R-O are similar for the same particle size range because significant volatile release occurs after the reducing zone. For the various particle sizes studied, gas temperature in O has a first order effect on single particle ignition, reducing ignition times in an 1800 K environment by a factor up to 5 times over those in 1300 K environment.

The next study on single particle ignition involves statistical ignition analysis aimed at investigating the ignition behavior of coal particles over a wide particle size range. Typically, coal particles are classified into narrow size ranges before single particle experiments. The ignition behavior of the narrow particle size range is then based on data from a few sampled particles, usually with no knowledge of what particle sizes in the size range are sampled. To address this limitation and eliminate the need for particle classification, single particle ignition experiments are conducted using 75-149 µm particles, and a new algorithm is developed that allows for simultaneous determination of particle size before ignition and ignition delay time from the experimental data. Using the new algorithm, statistical ignition data from these experiments are compared to the experimental results from ignition studies using four different narrow particle size ranges. The results show that the use of narrow size ranges in evaluating ignition behavior of coal compares well on average to the statistical ignition data. As expected, the results using the four narrow size ranges do not capture the variations in ignition delay times.

Apart from the single particle ignition experiments, the effects of particle size and the reducing-to-oxidizing environment on pulverized coal stream ignition are also studied. The results show that ignition delay time is independent of particle size at 1800 K gas temperatures in either the oxidizing or the reducing-to-oxidizing environment condition, but depends significantly on particle size at 1300 K. This finding indicates that ignition delay time is determined by the amount of volatiles released (depending on particle size), by the gas temperature, and by the local oxygen concentration. For all particle size ranges considered, homogeneous-to-heterogeneous ignition is observed. The reducing-to-oxidizing environment increases ignition delay times by 100% on average over those of oxidizing conditions at 1800 K gas temperature. In the reducing-to-oxidizing environment, longer ignition delay times are due to the oxidation of volatiles released before the particles exit the volatile flame front at the flame front. Hence, for homogeneous ignition to occur, more volatiles must be generated downstream of the reducing-to-oxidizing environment zone. On the other hand, at a 1300 K nominal gas temperature, ignition delay times and modes strongly depend on particle size in the stream. Interestingly though, there exists a crossover at which the ignition delay times in the reducing-to-oxidizing environment are less than those of the oxidizing environment for large particles (i.e., 90-124 µm and 125-149 µm) at 1300 K.

The experimental observations are consistent with model predictions. With both experimental and computational modeling tools, this dissertation offers new fundamental understanding of coal ignition as an early-stage process of coal combustion.


English (en)


Richard L. Axelbaum

Committee Members

Pratim Biswas, Rajan Chakrabarty, Benjamin M. Kumfer, Patricia Weisensee,


Permanent URL: