Combustion Chemistry of Fuels: Quantitative Speciation Data Obtained from an Atmospheric High-temperature Flow Reactor with Coupled Molecular-beam Mass Spectrometer

Summary

An investigation of the oxidative combustion chemistry of novel biofuels, fuel components, or jet fuels by comparison of quantitative speciation data is presented. The data can be used for kinetic model validation and enables fuel assessment strategies.This manuscript describes the atmospheric high-temperature flow reactor and demonstrates its capabilities.

Abstract

This manuscript describes a high-temperature flow reactor experiment coupled to the powerful molecular beam mass spectrometry (MBMS) technique. This flexible tool offers a detailed observation of chemical gas-phase kinetics in reacting flows under well-controlled conditions. The vast range of operating conditions available in a laminar flow reactor enables access to extraordinary combustion applications that are typically not achievable by flame experiments. These include rich conditions at high temperatures relevant for gasification processes, the peroxy chemistry governing the low temperature oxidation regime or investigations of complex technical fuels. The presented setup allows measurements of quantitative speciation data for reaction model validation of combustion, gasification and pyrolysis processes, while enabling a systematic general understanding of the reaction chemistry. Validation of kinetic reaction models is generally performed by investigating combustion processes of pure compounds. The flow reactor has been enhanced to be suitable for technical fuels (e.g. multi-component mixtures like Jet A-1) to allow for phenomenological analysis of occurring combustion intermediates like soot precursors or pollutants. The controlled and comparable boundary conditions provided by the experimental design allow for predictions of pollutant formation tendencies. Cold reactants are fed premixed into the reactor that are highly diluted (in around 99 vol% in Ar) in order to suppress self-sustaining combustion reactions. The laminar flowing reactant mixture passes through a known temperature field, while the gas composition is determined at the reactors exhaust as a function of the oven temperature. The flow reactor is operated at atmospheric pressures with temperatures up to 1,800 K. The measurements themselves are performed by decreasing the temperature monotonically at a rate of -200 K/h. With the sensitive MBMS technique, detailed speciation data is acquired and quantified for almost all chemical species in the reactive process, including radical species.

Introduction

Understanding combustion processes in the wake of modern, low-emission fuels from renewable resources is a challenge for today's societies' ecological and economic topics. They have the potential to reduce our dependence on fossil fuels, offset CO₂ emissions, and have a positive impact on harmful pollutant emissions such as soot and its toxic precursors¹. Combining this fast growing field with their utilization in modern combustor systems, the demand on a fundamental understanding of the governing chemical and physical processes has increased dramatically². Even today, the complex chemical reaction networks resulting from the radical chain reactions are still not fully understood. To analyze or even control phenomena like pollutant formation or (auto) ignition processes, the detailed knowledge of chemical reaction networks is a crucial piece of the puzzle³.

To investigate and understand those chemical reaction networks, experimental and numerical approaches are mandatory. Experimentally, the combustion chemistry is typically studied by applying experiments with simplified and well controlled flow environments to target specific questions. The high complexity and dynamics of individual sub processes prevent exact reproduction of the conditions of technical combustors by the fundamental experiments, while allowing the tracking of the designated key features such as temperature, pressure, heat release, or chemical species. Early on, the need for different experimental approaches became apparent, each tackling a specific question and providing a subsequent set of information contributing to the overall global picture of the combustion chemistry. To cover the full range of conditions and gather those subsequent information sets to describe complex conditions occurring in technical systems various approaches were successfully developed. Well established techniques include:

• Shock tubes^4,5,6 and rapid compression machines⁷. These devices provide high control of pressure and temperature over a wide range. However, the accessible reaction time and suitable analytical techniques are limited.
• Laminar premixed flames^3,8,9,10,11 are ideal to gain high-temperature conditions in combination with a simple flow field. Since the spatial dimension of the reaction zone decreases with increasing pressure, premixed flames are typically investigated at low-pressure conditions for speciation purposes.
• Counterflow diffusion flames^12,13,14,15 are ideal for investigating the flamelet regime in turbulent combustion. They mimic the strain due to inhomogeneities in a real turbulent flow, but are, again, highly limited in analytical speciation techniques.
• Various reactor experiments^16,17,18 (static, stirred and plug-flow) provide access to high-pressure environments, while temperatures are typically lower compared to flame environments. Common approaches are:
  • Static reactors are widely used for e.g. pulse photolysis experiments, but are in general limited by long residence times and low temperatures.
  • Jet-stirred reactors, i.e. gas version of a perfectly stirred reactor (PSR), rely on the efficient mixing of the gas phase and can be operated at steady state with constant residence time, temperature and pressure, making it easy to model. However, molecules have time to migrate to the hot surfaces and undergo heterogeneous reactions.
  • Numerous flow reactor approaches are known, with the plug flow reactor (PFR) as one of the most popular approaches for describing chemical reactions in continuous, flowing systems of cylindrical geometry. Plug flow conditions at steady state are assumed with fixed residence time of the plug as a function of its position for ideal PFRs.

Complementary to those valuable techniques in the field of experimental combustion kinetics, a high-temperature laminar flow reactor experiment^19,20 employing the molecular beam mass spectrometry (MBMS) technique for tracing species development in detail is presented^21,22 herein. Laminar flow conditions, working at atmospheric pressure and accessible temperatures up to 1,800 K are the main characteristics of the flow reactor, while the sensitive MBMS technique allows the detection of almost all chemical species present in the combustion process. This includes highly reactive species such as radicals that are not or hardly traceable with other detection methods. The MBMS technique is widely used for the detailed investigation of reaction networks in flames of conventional and modern alternative fuels, such as alcohols or ethers^23,24,25 and has demonstrated to be of great value for modern kinetic model development.

Figure 1 shows the schematic of the high-temperature flow reactor with a zoomed frame of the sampling probe (A) and two pictures highlighting the overall experiment (B) and the probe setup (C). The system can be divided in two segments: first, the high-temperature flow reactor with gas supplies and vaporizer system and second, the MBMS time-of-flight detection system. In operation, the exit of the flow tube is mounted directly to the sampling nozzle of the MBMS system. The gas is sampled directly from the reactor outlet and transferred to the high-vacuum detection system. Here, ionization is performed by electron ionization with subsequent time-of-flight detection.

The reactor has a 40-mm inner diameter ceramic (Al₂O₃) pipe of 1,497 mm length placed in a high temperature oven (e.g., Gero, Type HTRH 40-1000). The total heated section is 1,000 mm in length. Gases are fed premixed and pre-vaporized into the reactor by a tempered flange (typically tempered to ~80 °C). The highly diluted (ca. 99 vol% in Ar), laminar flowing reactant mixture passes through a known temperature profile (details on temperature characterization will be given below). Detection of the gas composition takes place at the reactor outlet as a function of the oven temperature. Measurements are performed at constant inlet mass flow, while a monotonically decreasing temperature ramp (-200 K/h) is applied to the oven in the range of 1,800 K to 600 K. Note that similar results may be obtained when distinct temperatures are measured at isothermal oven temperatures and thermal inertia is considered properly. The thermal stabilization of the system still takes some time and the temperature ramp is selected as a compromise of averaging time for a (negligible) small temperature increment and total measurement time per series. The averaging time (45 s) of the MBMS corresponds to 2.5 K. The resulting residence times are around 2 s (at 1,000 K) for the given conditions. Finally, due to the temperature reproducibility, a relative precision of the measured temperatures of ±5 K or better can be stated for the present reactor experiment.

Figure 2 shows the schematic of the vaporizing system, optimized to investigate even complex hydrocarbon mixtures such as technical jet fuels. All input streams are metered in high precision (accuracy ±0.5 %) by Coriolis mass flow meters. Vaporization of the fuel is realized by a commercial vaporizer system at temperatures up to 200 °C. All supply lines with pre-vaporized fuels are preheated with temperatures of typically 150 °C to prevent condensation of the liquid fuels, while avoiding thermal degradation at the same time. Complete and stable vaporization is routinely checked and may even occur at temperatures below the normal boiling point of the respective fuels. Complete evaporation was ensured by the small fuel fraction and the low partial pressure (typically below 100 Pa) needed.

The gases are sampled by a quartz cone at the centerline of the reactor exit at ambient pressures (around 960 hPa) as seen in more detail in the zoomed frame of Figure 1. The nozzle tip has a 50 μm orifice, which is located roughly 30 mm inside the ceramic tube at the end of the reaction zone. Note, that the sampling location is fixed with respect to the inlet. Thermal expansion of the oven tube only takes place at the outlet, which is not mechanically connected to the sampling system resulting in a temperature independent length of the reaction segment. All reactions are immediately quenched due to the formation of a molecular beam, when gasses are expanded into high vacuum (two differential pumping stages; 10^-2 and 10^-4 Pa)^25,26. The sample is guided to the ion source of an electron impact (EI) time-of-flight (TOF) mass spectrometer (mass resolution R = 3,000) capable of determining the exact mass of the present species in suitable precision to determine the elemental composition within a C/H/O system. The electron energy is set to low values (typically 9.5-10.5 eV) in order to minimize fragmentation due to the ionization process. Note that the diluent and reference species argon is still detectable due to the broad energy distribution of the ionizing electrons (1.4 eV FWHM). While Ar can be measured with good S/N, the low electron energy does not allow for sufficient determination of the major species (H₂O, CO₂, CO, H₂, O₂, and fuel) profiles, which are present in significant lower concentrations.

In addition to the detection by TOF, a residual gas analyzer (RGA), i.e. a quadrupole mass spectrometer, is placed in the ionization chamber to monitor the six species above with a higher electron energy (70 eV) simultaneously to the MBMS-TOF measurements.

Protocol

1. Setup of the molecular beam mass spectrometer (MBMS) and flow reactor system

1. Heat oven to designated start temperature, which is the highest temperature in designated measurement series. For typical conditions of Jet A-1 with Φ=1, total oxidation is observed below 850 °C (~1,100 K). The choice of proper starting temperatures depends on the chemical nature of the investigated fuel and the stoichiometry (Φ).
2. Prepare Time-of-Flight (TOF) spectrometer for intermediate species detection. The TOF spectrometer is aligned to the molecular beam and thus provides reliable detection of labile species.
   NOTE: The mass resolution is suitable for the determination of the elemental composition in a C/H/O system. To avoid fragmentation, chose soft ionization conditions. Values of typically 9.5 to 10.5 eV have been proven suitable for a typical investigation of combustion intermediates.
3. Prepare quadrupole spectrometer for major species detection.
   NOTE: Because the quadrupole spectrometer (also called Residual Gas Analyzer, RGA) is placed inside the ionization chamber of the MBMS system nearby the molecular beam, just the wall-scattered background gases are measured. Since major species are stable, the background pressure well reflects their concentration at the sampling tip. For better signal-to-noise ratio, chose a high ionization energy of 70 eV in this case.

2. Preparation of the fuel sample

1. Prepare the metal syringe for fuel supply.
   Caution: Use proper personal protective equipment for fuel handling.
   1. Fill 30 mL of the fuel sample to the vaporizers metal syringe.
   2. Pressurize fuel supply system (metal syringe) up to 5 bar by adding pressurized air to the system though opening the valve.
   3. Vent fuel lines and Coriolis mass flow meter by opening the valves in the fuel supply lines.
2. Heat up the vaporizer and gas supply lines. In the given setup, temperatures far below the normal pressure boiling point can be applied due to the high dilution. Ensure that the vapor pressure at the designated temperature of the highest boiling compound of the fuel is higher than its partial pressure in the gas stream. Typically, 200 °C is adequate for Jet A-1.
3. Note that the coldest spot in the system is the tempered inlet flange to the oven. Make sure the diluted fuel cannot recondense at this spot. For typical Jet A-1 (99% dilution) setting the water cooling system to 80 °C is adequate.

3. Measurement and data acquisition

1. Place oven to the sampling position. The sampling cone must be placed inside the oven's ceramic tube. In the present experiment, the sampling location is close to the plateau value of the spatial temperature profile of the oven.
   NOTE: The picture in Figure 1 shows the cone and the reactor. The heated reactor (blue) is moved on rails towards the quartz cone.
2. Start diluent of choice by adding gas through the Coriolis mass flow meter.
   NOTE: The mass flow meters are controlled by the original software package. Here, mass flow values can be set. Typically, 99% argon is used. In general, for liquid fuels, the diluent flow may be split to vaporizer and oxidizer gas streams by using an additional Coriolis mass flow meter parallel to Cori Flow 2 passing the heated line and connected directly to the oxidizer stream from Cori Flow 3.
3. Start continuous data recording (TOF and Quadrupole) by using the designated instrumental software.
   NOTE: Click Start-Button in Quadrupole software. Click Start-Button in TOF software.
   1. Add oxidizer O₂ by setting the appropriate flow condition of the Coriolis flow meter software. Observe the incoming oxidizer as new peak in mass spectrum.
4. Add fuel by setting the appropriate flow condition of the Coriolis mass flow meter.
   1. Check spectra. If complete oxidation in case of lean and stoichiometric conditions is achieved, observe a stable CO₂ mass signal.
5. When signal intensities are stable for 4 to 5 measurements, a stabilization period ends. After the stabilization period, apply the continuous temperature decay ramp of typically -200 K/h to the oven. This leads to typical measurement times of 2 h per measurement run.
   1. At a specific oven temperature during the ramp, observe a rapid change of the mass spectra. Sole combustion products (H₂O, CO₂ and CO, H₂ in rich cases) start to disappear and small combustion intermediates become detectable.
      NOTE: With further decreasing temperature, visible intermediates become larger and larger. At cold oven temperatures, only the signal of fuel compounds and oxygen can be observed; no reactions are taking place within the present residence time of the reactor.
6. When the final temperature is stabilized (typically 500 °C; 10 min), switch off the oxidizer.
   1. Continue recording measurements. Obtain fuel characterization (composition of fragmentation) measurements at conditions without oxidizer.
7. Switch off fuel in the Coriolis mass flow meter software by setting the value to 0. Data will be still recorded; use these spectra for background measurement.
8. Stop data recording by clicking the stop button in the software.

4. Calibration measurements

1. For calibration issues, mount a closed chamber in front of the sampling cone.
   NOTE: The closed chamber is a tube, which is placed in front of the nozzle by hand.
2. Open valve to pump. The chamber is evacuated.
3. Apply binary mixtures (hydrocarbon of interest and Ar) or commercial calibration gasses for calibration. For demonstration, a customized gas mixture with CO and CO₂ and argon is used here.
4. Adjust the pressure in the calibration chamber by a needle valve to obtain signal intensity above the signal-to-noise ratio and below the saturation limit.
5. Start calibration measurements by starting data recording as performed for the individual measurement in 3.5 by clicking the start button in the TOF software.

5. Data processing

NOTE: Recoded spectra must be matched to the respective oven temperature they are recorded at.

1. At each recorded temperature, for each chosen species, calculate its mole fraction from the corresponding signal. Plot the mole fraction profiles vs. oven temperature (Figure 3).

Results

A typical mass spectrum of the sampled gas composition is shown in Figure 3. With the given setup of a mass resolution of approx. 3,000, species up to m/z = 260 u can be detected within the C/H/O system. After a mass calibration procedure, the peaks are integrated for each mass-to-charge (m/z) ratio with deconvolution algorithms for evaluating under-resolved signals. After background and fragmentation corrections, the signal can be quantified using the approp...

Discussion

The presented combination of an atmospheric high-temperature flow reactor with a molecular-beam mass spectrometry detection system enables quantitative speciation data for a range of operating conditions. Several studies^21,22,23,27 demonstrated the flexibility of the experiment starting from rich methane conditions relevant for partial oxidation phenomena (φ = 2.5), to investigating the co...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Time-Of-Flight MBMS	Kaesdorf	n.a.	custom design
Molecular Beam Samling Interface	self made	n.a.	custom design
Laminar Flow Reactor	Gero	Type HTRH 40-1000	custom design
Quadrupole MS	Hiden	HAL/3F 301	adapted to ionization chamber
Vaporizer	Bronkhorst	CEM	Vaporizer
Mass Flow Meter	Bronkhorst	Mini Cori-Flow M12, M13, M14	Flow Controller
Jet A-1	n.a.	n.a.	Standard Jet fuel of interest
Metal syringe	Hugo Sachs	70-2252	Fuel Supply
Heating Hoses	Hillesheim	HMI series	Gas Preheating
Gas	Linde	Ar, O2	Diluent, Oxidizer