The experimental setup of time-resolved synchrotron x-ray diffraction to probe the structural and chemical changes of thin metal foils in combustion reaction initiated by electrical pulse heating. The setup utilizes an intense monochromatic x-rays from the APS, a 2D pixel array x-ray detector (PILATUS), and a custom-designed fast-rotating chopper. The lower left images depict the microscopic views of the Zr sample configuration (left) and during combustion (right). 

Contexts in source publication

Context 1

... metals with high oxygen affinities such as Zr, Ti, and Fe can rapidly react with oxygen at high temperatures, yielding a wide range of transition metal oxides. These highly exothermic, self-fueling metal oxidation reactions (or metal combustions) are complex and strongly depend on microstructures, crystal structures, melting temperatures, and oxygen diffusivities of metals and metal oxides. The presence of solid-solid or solid-liquid phase transitions, for example, has a dramatic effect on the ignition process and combustion kinetics, which can lead to a micro-explosion or an interesting “spear point” formation of oxide products. 1 Yet, a wide range of oxidation states viable in these transition metals can give rise to an array of different oxides depending on temperatures and oxygen concentrations and, thereby, greatly complicate the reaction pathways. Further- more, the dynamic (or transient) nature of combustions may result in the path-dependent phase and chemical changes and quench metastable structures and secondary products. Therefore, obtaining time-resolved structural data during metal ignitions and combustions are critical to understanding the chemical mechanisms, energetics, and kinetics. The most critical process governing metal combustions in air is oxygen diffusion, which strongly depends on the microstructure of metal surface. The presence of thin oxide layers on the surface of small metal particles limits the diffusion of oxygen molecules and thus the combustion in solid metals. As a result, the combustion occurs typically from molten phases of metals above the melting temperatures. 2,3 In contrast, small oxygen-free metal particles (sub- to a few l m in diameter) freshly produced from metallic glass and lamellae metal alloys with nm-scale microstructures can ignite at relatively low temperatures (well below the melting temperatures of metals) and burn nearly completely within very short reaction time ( < 1 ms) to produce highly oxidized products. 4 Because of large oxygen diffusion barrier of metal oxides and high thermal conductivity of metals, the oxidation reactions of larger metal particles (say 10–100 l m) in surrounding air are typically limited to a very thin layer (less than a few l m) — forming huge gas diffusion and reaction barriers and, thus, leaving a significant portion of large frag- ments unreacted. The delicate balance among thermal (heat) conduction, oxygen (mass) diffusion, and micro (surface) structures is central to controlling chemical mechanisms, energetics, and dynamics of metal combustions. This again underscores the significance of time-resolved thermal and structural information over the relevant time (1 l s to 1 ms) and spatial (atomistic) scales. The present phase and chemical analysis of metal combustions are primarily based on the temperature profile, phase diagram and postburn microprobe composition analysis 5,6 — not on the real-time structural information. This is unfortunate considering the fact that synchrotron x-ray diffraction is capable of probing structural changes in exothermic solid/solid reactions including self-propagating high-temperature reactions, 7–9 intermetallic reactions of mul- tilayer nanofoils, 10,11 and metathesis reactions. 12,13 It is pre- sumably due to a substantially faster time resolution (in l s) required for probing metal/gas reactions. Recently, we have developed a relatively simple in situ time- and angle- resolved x-ray diffraction (TARXD) technique to measure the structural and chemical evolutions associated with rapidly propagating intermetallic reactions in a l s time- resolution. 14 In this study, we have applied this TARXD along with the fast time-resolved temperature measurements, probing real-time structural and chemical evolutions associated with oxygen-diffusion limited metal combustions in bulk metallic foils of Zr, Ti, and Fe. Metal foils (99.8% or better purity) of Zr and Ti (25 l m in thickness from Alfa Aesar) and Fe (10 l m from Goodfellow) were used for the samples. These metal foils were cut into strips of about 0.15 mm wide and 2–3 mm long for combustion and time-resolved x-ray diffraction experiments (as shown in Fig. 1). The metal strip sample was placed on a nonconductive ceramic substrate and anchored down with copper electrodes at the both ends. The spacing between the two electrodes was about 0.3–0.5 mm. Along with the metal strip thickness and width, it defines an active combustion volume of about 0.5–2.0 times 10 À 3 mm 3 . The electrical resistance of metal strips was in the range of 0.2–0.4 X measured with a digital multimeter. Underneath the metal strip, a small through-hole of 0.2 mm in diameter was drilled for the incoming x-ray beam. A DC current source (with a 5 A/6 V capacity) was connected to the copper electrodes. The circuit opening and closure were achieved with an electrical relay switch. The electrical pulse from the digital delay generator (SRS DG645) was used as a gate control signal to trigger the electrical relay and initiate metal combustion. Upon pulsed electrical heating, the metal strip quickly reacts with oxygen in air resulting in highly exothermic reactions, emitting intense thermal radiations. The representative snapshot of Zr combustion recorded on a high-speed camera (Photron FastCam SA.1) is shown on the lower left in Fig. 1. A digital phosphor oscilloscope (Tektronix DPO2000 with 100–200 MHz bandwidth) was used to monitor the voltage evolution of the reacting sample. We used a 6-channel optical pyrometer to measure the temperature evolution of metal strip in combustion, based on the Planck’s law. 15,16 The optical pyrometer was constructed, for each channel, using a photomultiplier tube (Hamamatsu H7732P-11 PMT with a rise time of 2.2 ns), a narrow spectral band-pass filter ( D k 1⁄4 10 nm at FWHM in the range of 550–800 nm with an increment of 50 nm), and a set of neutral density (ND) filters to adjust the incident light intensity within the linear response of PMTs. The incident light was transmit- ted through six optical fibers (each 0.2 mm core diameter) bundled in a hexagonal pattern at the probe side. The PMT signal was recorded onto two oscilloscopes (Tektronix DPO2000) with 50 X terminations, matching the cable impedance and thereby avoiding signal distortions from reflections. Before measurements, the entire pyrometer system, including PMTs, fibers, filters and other optics used, was calibrated with a black- body radiation source (BBRS, OL 480 from Optronic Labora- tories). Time-resolved temperatures were obtained by fitting the measured emission intensities at six discrete wavelengths to a gray body radiation equation. The uncertainty of dynamic temperatures measured by the present optical pyrometer varies is in the range of 3–8%, primarily centered around 5%. X-ray diffraction experiments were performed using intense monochromatic ( k 1⁄4 0.8638 A ̊ ) x-rays from the 16IDD/HPCAT beamline at the Advanced Photon Source (APS). A 2D pixel array detector (PILATUS 100 K) was used to record the TARXD image, which had an active area of 83.8 Â 33.5 mm 2 with 487 Â 195 pixels (each pixel size of 172 Â 172 l m 2 ). A fast-rotating metal disk (150 mm in diameter) with four orthogonal opening slots of 1.56 degrees was used to chop the diffraction beam into different time domains. In this configuration, a quarter of the chopper disk covers the entire detector area. The chopper was placed in front of the detector with a few millimeters spacing. The incident beam was aligned to the center of the chopper disk along the vertical center of the detector. As the disk rotates, the slot sweeps across the entire detector active area clockwise along the Debye-Scherrer’s rings. As a result, small portions of diffraction rings are recorded at different azimuth angles on the detector at different time. Thus, the diffraction beam is dispersed as a function of time, recording the structural evolution of reacting samples. As the diffraction arcs are recorded continuously, there is no down (or blackout) time in the entire 2D TARXD image representing about 3 ms long event – the critical initial stage in combustions. The time-resolution is then determined by the chopper speed and the slot width. We used a time-resolution of $ 45 l s in this work. Utilizing the kinetic mode of the detector, multiple frames can be recorded over several hundreds of ms with a relatively small readout time of 3 ms between the frames. This extended record time is important to probe the slow chemical and structural changes occurring later times in metal combustions. More detailed description of the time- resolved x-ray diffraction can be found elsewhere. 14,17 Figure 2 illustrates the measured (in black) and smoothed (red) time-resolved temperatures, as well as voltage (green) changes of Zr foil during an electric pulse heating (blue). Because of the limited spectral response of PMTs in the range of 185–850 nm, the lower temperature cut-off of the present pyrometer is $ 1100–1200 K. Based on the rising slope of the measured time-resolved temperatures, Zr metal combusts at the heating rate of $ 1.5*10 6 K/s. ...

Context 2

... of metals and metal oxides. The presence of solid-solid or solid-liquid phase transitions, for example, has a dramatic effect on the ignition process and combustion kinetics, which can lead to a micro-explosion or an interesting “spear point” formation of oxide products. 1 Yet, a wide range of oxidation states viable in these transition metals can give rise to an array of different oxides depending on temperatures and oxygen concentrations and, thereby, greatly complicate the reaction pathways. Further- more, the dynamic (or transient) nature of combustions may result in the path-dependent phase and chemical changes and quench metastable structures and secondary products. Therefore, obtaining time-resolved structural data during metal ignitions and combustions are critical to understanding the chemical mechanisms, energetics, and kinetics. The most critical process governing metal combustions in air is oxygen diffusion, which strongly depends on the microstructure of metal surface. The presence of thin oxide layers on the surface of small metal particles limits the diffusion of oxygen molecules and thus the combustion in solid metals. As a result, the combustion occurs typically from molten phases of metals above the melting temperatures. 2,3 In contrast, small oxygen-free metal particles (sub- to a few l m in diameter) freshly produced from metallic glass and lamellae metal alloys with nm-scale microstructures can ignite at relatively low temperatures (well below the melting temperatures of metals) and burn nearly completely within very short reaction time ( < 1 ms) to produce highly oxidized products. 4 Because of large oxygen diffusion barrier of metal oxides and high thermal conductivity of metals, the oxidation reactions of larger metal particles (say 10–100 l m) in surrounding air are typically limited to a very thin layer (less than a few l m) — forming huge gas diffusion and reaction barriers and, thus, leaving a significant portion of large frag- ments unreacted. The delicate balance among thermal (heat) conduction, oxygen (mass) diffusion, and micro (surface) structures is central to controlling chemical mechanisms, energetics, and dynamics of metal combustions. This again underscores the significance of time-resolved thermal and structural information over the relevant time (1 l s to 1 ms) and spatial (atomistic) scales. The present phase and chemical analysis of metal combustions are primarily based on the temperature profile, phase diagram and postburn microprobe composition analysis 5,6 — not on the real-time structural information. This is unfortunate considering the fact that synchrotron x-ray diffraction is capable of probing structural changes in exothermic solid/solid reactions including self-propagating high-temperature reactions, 7–9 intermetallic reactions of mul- tilayer nanofoils, 10,11 and metathesis reactions. 12,13 It is pre- sumably due to a substantially faster time resolution (in l s) required for probing metal/gas reactions. Recently, we have developed a relatively simple in situ time- and angle- resolved x-ray diffraction (TARXD) technique to measure the structural and chemical evolutions associated with rapidly propagating intermetallic reactions in a l s time- resolution. 14 In this study, we have applied this TARXD along with the fast time-resolved temperature measurements, probing real-time structural and chemical evolutions associated with oxygen-diffusion limited metal combustions in bulk metallic foils of Zr, Ti, and Fe. Metal foils (99.8% or better purity) of Zr and Ti (25 l m in thickness from Alfa Aesar) and Fe (10 l m from Goodfellow) were used for the samples. These metal foils were cut into strips of about 0.15 mm wide and 2–3 mm long for combustion and time-resolved x-ray diffraction experiments (as shown in Fig. 1). The metal strip sample was placed on a nonconductive ceramic substrate and anchored down with copper electrodes at the both ends. The spacing between the two electrodes was about 0.3–0.5 mm. Along with the metal strip thickness and width, it defines an active combustion volume of about 0.5–2.0 times 10 À 3 mm 3 . The electrical resistance of metal strips was in the range of 0.2–0.4 X measured with a digital multimeter. Underneath the metal strip, a small through-hole of 0.2 mm in diameter was drilled for the incoming x-ray beam. A DC current source (with a 5 A/6 V capacity) was connected to the copper electrodes. The circuit opening and closure were achieved with an electrical relay switch. The electrical pulse from the digital delay generator (SRS DG645) was used as a gate control signal to trigger the electrical relay and initiate metal combustion. Upon pulsed electrical heating, the metal strip quickly reacts with oxygen in air resulting in highly exothermic reactions, emitting intense thermal radiations. The representative snapshot of Zr combustion recorded on a high-speed camera (Photron FastCam SA.1) is shown on the lower left in Fig. 1. A digital phosphor oscilloscope (Tektronix DPO2000 with 100–200 MHz bandwidth) was used to monitor the voltage evolution of the reacting sample. We used a 6-channel optical pyrometer to measure the temperature evolution of metal strip in combustion, based on the Planck’s law. 15,16 The optical pyrometer was constructed, for each channel, using a photomultiplier tube (Hamamatsu H7732P-11 PMT with a rise time of 2.2 ns), a narrow spectral band-pass filter ( D k 1⁄4 10 nm at FWHM in the range of 550–800 nm with an increment of 50 nm), and a set of neutral density (ND) filters to adjust the incident light intensity within the linear response of PMTs. The incident light was transmit- ted through six optical fibers (each 0.2 mm core diameter) bundled in a hexagonal pattern at the probe side. The PMT signal was recorded onto two oscilloscopes (Tektronix DPO2000) with 50 X terminations, matching the cable impedance and thereby avoiding signal distortions from reflections. Before measurements, the entire pyrometer system, including PMTs, fibers, filters and other optics used, was calibrated with a black- body radiation source (BBRS, OL 480 from Optronic Labora- tories). Time-resolved temperatures were obtained by fitting the measured emission intensities at six discrete wavelengths to a gray body radiation equation. The uncertainty of dynamic temperatures measured by the present optical pyrometer varies is in the range of 3–8%, primarily centered around 5%. X-ray diffraction experiments were performed using intense monochromatic ( k 1⁄4 0.8638 A ̊ ) x-rays from the 16IDD/HPCAT beamline at the Advanced Photon Source (APS). A 2D pixel array detector (PILATUS 100 K) was used to record the TARXD image, which had an active area of 83.8 Â 33.5 mm 2 with 487 Â 195 pixels (each pixel size of 172 Â 172 l m 2 ). A fast-rotating metal disk (150 mm in diameter) with four orthogonal opening slots of 1.56 degrees was used to chop the diffraction beam into different time domains. In this configuration, a quarter of the chopper disk covers the entire detector area. The chopper was placed in front of the detector with a few millimeters spacing. The incident beam was aligned to the center of the chopper disk along the vertical center of the detector. As the disk rotates, the slot sweeps across the entire detector active area clockwise along the Debye-Scherrer’s rings. As a result, small portions of diffraction rings are recorded at different azimuth angles on the detector at different time. Thus, the diffraction beam is dispersed as a function of time, recording the structural evolution of reacting samples. As the diffraction arcs are recorded continuously, there is no down (or blackout) time in the entire 2D TARXD image representing about 3 ms long event – the critical initial stage in combustions. The time-resolution is then determined by the chopper speed and the slot width. We used a time-resolution of $ 45 l s in this work. Utilizing the kinetic mode of the detector, multiple frames can be recorded over several hundreds of ms with a relatively small readout time of 3 ms between the frames. This extended record time is important to probe the slow chemical and structural changes occurring later times in metal combustions. More detailed description of the time- resolved x-ray diffraction can be found elsewhere. 14,17 Figure 2 illustrates the measured (in black) and smoothed (red) time-resolved temperatures, as well as voltage (green) changes of Zr foil during an electric pulse heating (blue). Because of the limited spectral response of PMTs in the range of 185–850 nm, the lower temperature cut-off of the present pyrometer is $ 1100–1200 K. Based on the rising slope of the measured time-resolved temperatures, Zr metal combusts at the heating rate of $ 1.5*10 6 K/s. Extrapolating the heating rate to ambient temperature yields an induction time of less than 0.1 ms. The combustion temperature reaches the maximum at $ 3200 K (defined as the burst temperature or T B ), which is consistent with the literature values. 18–20 This burst temperature is higher than the melting temperatures of both Zr (2128 K) and ZrO 2 (2983 K) indicated by dashed lines in Fig. 2. Figure 3 shows several TARXD images recorded at a time-resolution of $ 45 l s, each representing 3 ms-long structural and chemical changes of a Zr foil during the combustion process. Time runs clockwise. There is a 3 ms blackout time between the images. The onset of the electrical pulse heating is marked by “ t o ” with a straight line in Fig. 3(b). The polymorphous transition from hcp a -Zr to bcc b -Zr is also marked with another straight line on the right. A small segment of a white stripe along the radial direction of azimuth in Figs. 3(b)–3(e) represents the doubly exposed area of the diffraction signal from the next incoming slot of the chopper wheel, which is used to locate the onset time in the images. Figure 4 shows the “caked” TARXD images of Fig. 3, ...