Volume 34 Issue 1
Feb.  2021
Turn off MathJax
Article Contents

Guo-dong Zhang, Li-chang Guan, Zi-feng Yan, Min Cheng, Hong Gao. A Three-Dimensional Velocity-Map Imaging Setup Designed for Crossed Ion-Molecule Scattering Studies[J]. Chinese Journal of Chemical Physics , 2021, 34(1): 71-80. doi: 10.1063/1674-0068/cjcp2012219
Citation: Guo-dong Zhang, Li-chang Guan, Zi-feng Yan, Min Cheng, Hong Gao. A Three-Dimensional Velocity-Map Imaging Setup Designed for Crossed Ion-Molecule Scattering Studies[J]. Chinese Journal of Chemical Physics , 2021, 34(1): 71-80. doi: 10.1063/1674-0068/cjcp2012219

A Three-Dimensional Velocity-Map Imaging Setup Designed for Crossed Ion-Molecule Scattering Studies

doi: 10.1063/1674-0068/cjcp2012219
More Information
  • Corresponding author: Gao Hong, E-mail: honggao2017@iccas.ac.cn
  • These authors contributed equally to this work.
  • Received Date: 2020-12-27
  • Accepted Date: 2021-01-18
  • Publish Date: 2021-02-27
  • In this study, we report the design and simulation of an electrostatic ion lens system consisting of 22 round metal plates. The opening of the extractor plate is covered with metal mesh, which is for shielding the interaction region of the lens system from the high DC voltages applied to all other plates than the repeller and extractor plates. The Simion simulation shows that both velocity-mapping and time focusing can be achieved simultaneously when appropriate voltages are applied to each of the plates. This makes the ion lens system be able to focus large ionic volumes in all three dimensions, which is an essential requirement for crossed ion-molecule scattering studies. A three-dimensional ion velocity measurement system with multi-hit and potential multi-mass capability is built, which consists of a microchannel plate (MCP), a P47 phosphor screen, a CMOS camera, a fast photomultiplier tube (PMT), and a high-speed digitizer. The two velocity components perpendicular to the flight axis are measured by the CMOS camera, and the time-of-flight, from which the velocity component along the flight axis can be deduced, is measured by the PMT. A Labview program is written to combine the two measurements for building the full three-dimensional ion velocity in real time on a frame-by-frame basis. The multi-hit capability comes from the fact that multiple ions from the camera and PMT in the same frame can be correlated with each other based on their various intensities. We demonstrate this by using the photodissociation of CH3I at 304 nm.
  • These authors contributed equally to this work.
  • 加载中
  • [1] V. G. Anicich, Astrophys. J. Supplement Series 84, 215 (1993).
    [2] P. B. Armentrout, in Advances In Atomic, Molecular, and Optical Physics, Benjamin Bederson and Herbert Walther Eds., San Diego: Academic Press, 187 (2000).
    [3] A. B. Fialkov, Prog. Energy Combust. Sci. 23, 399 (1997).
    [4] C. H. DePuy and V. M. Bierbaum, Acc. Chem. Res. 14, 146 (1981). doi:  10.1021/ar00065a003
    [5] W. Lindinger, A. Hansel, and A. Jordan, Int. J. Mass Spectrom. Ion Proc. 173, 191 (1998). doi:  10.1016/S0168-1176(97)00281-4
    [6] S. T. Arnold, J. V. Seeley, J. S. Williamson, P. L. Mundis, and A. A. Viggiano, J. Phys. Chem. A 104, 5511 (2000).
    [7] A. A. Viggiano and R. A. Morris, J. Phys. Chem. 100, 19227 (1996). doi:  10.1021/jp962084x
    [8] E. Haufler, S. Schlemmer, and D. Gerlich, J. Phys. Chem. A 101, 6441 (1997). doi:  10.1021/jp9707246
    [9] L. A. Angel and K. M. Ervin, J. Am. Chem. Soc. 125, 1014 (2003). doi:  10.1021/ja021003+
    [10] K. Rempala and K. M. Ervin, J. Chem. Phys. 112, 4579 (2000). doi:  10.1063/1.481016
    [11] C. Y. Ng, J. Phys. Chem. A 106, 5953 (2002). doi:  10.1021/jp020055i
    [12] Y. C. Chang, Y. T. Xu, Z. Lu, H. Xu, and C. Y. Ng, J. Chem. Phys. 137, 104202 (2012). doi:  10.1063/1.4750248
    [13] Y. T. Xu, B. Xiong, Y. C. Chang, and C. Y. Ng, J. Chem. Phys. 137, 241101 (2012). doi:  10.1063/1.4773099
    [14] Y. P. Chang, K. Długołȩcki, J. Küpper, D. Rösch, D. Wild, and S. Willitsch, Science 342, 98 (2013). doi:  10.1126/science.1242271
    [15] A. Kilaj, H. Gao, D. Rösch, U. Rivero, J. Küpper, and S. Willitsch, Nat. Commun. 9, 2096 (2018). doi:  10.1038/s41467-018-04483-3
    [16] L. S. Petralia, A. Tsikritea, J. Loreau, T. P. Softley, and B. R. Heazlewood, Nat. Commun. 11, 173 (2020). doi:  10.1038/s41467-019-13976-8
    [17] L. S. Pei, E. Carrascosa, N. Yang, S. Falcinelli, and J. M. Farrar, J. Phys. Chem. Lett. 6, 1684 (2015). doi:  10.1021/acs.jpclett.5b00517
    [18] J. Mikosch, S. Trippel, C. Eichhorn, R. Otto, U. Lourderaj, J. X. Zhang, W. L. Hase, M. Weidemüller, and R. Wester, Science 319, 183 (2008). doi:  10.1126/science.1150238
    [19] R. Wester, Phys. Chem. Chem. Phys. 16, 396 (2014).
    [20] E. Carrascosa, J. Meyer, and R. Wester, Chem. Soc. Rev. 46, 7498 (2017). doi:  10.1039/C7CS00623C
    [21] J. Hu, C. X. Wu, Y. S. Ma, and S. X. Tian, J. Phys. Chem. A 122, 9171 (2018). doi:  10.1021/acs.jpca.8b08005
    [22] C. X. Wu, J. Hu, M. M. He, and S. X. Tian, J. Phys. Chem. A 123, 8536 (2019). doi:  10.1021/acs.jpca.9b06607
    [23] C. X. Wu, J. Hu, M. M. He, Y. Y. Zhi, and S. X. Tian, Phys. Chem. Chem. Phys. 22, 4640 (2020). doi:  10.1039/C9CP06289K
    [24] A. T. J. B. Eppink and D. H. Parker, Rev. Sci. Instrum. 68, 3477 (1997). doi:  10.1063/1.1148310
    [25] J. J. Lin, J. G. Zhou, W. C. Shiu, and K. P. Liu, Rev. Sci. Instrum. 74, 2495 (2003). doi:  10.1063/1.1561604
    [26] S. R. Yu, D. F. Yuan, W. T. Chen, X. M. Yang, and X. A. Wang, J. Phys. Chem. A 119, 8090 (2015). doi:  10.1021/acs.jpca.5b04438
    [27] S. Trippel, M. Stei, R. Otto, P. Hlavenka, J. Mikosch, C. Eichhorn, U. Lourderaj, J. X. Zhang, W. L. Hase, M. Weidemüller, and R. Wester, J. Phys. : Confer. Ser. 194, 012046 (2009). doi:  10.1088/1742-6596/194/1/012046
    [28] J. Hu, C. X. Wu, and S. X. Tian, Rev. Sci. Instrum. 89, 066104 (2018). doi:  10.1063/1.5026822
    [29] S. K. Lee, F. Cudry, Y. F. Lin, S. Lingenfelter, A. H. Winney, L. Fan, and W. Li, Rev. Sci. Instrum. 85, 123303 (2014). doi:  10.1063/1.4903856
    [30] C. Weeraratna, C. Amarasinghe, S. K. Lee, W. Li, and A. G. Suits, J. Chem. Phys. 149, 084202 (2018). doi:  10.1063/1.5040589
    [31] W. C. Wiley and I. H. McLaren, Rev. Sci. Instrum. 26, 1150 (1955). doi:  10.1063/1.1715212
    [32] A. I. Chichinin, K. H. Gericke, S. Kauczok, and C. Maul, Int. Rev. Phys. Chem. 28, 607 (2009). doi:  10.1080/01442350903235045
    [33] W. Li, S. D. Chambreau, S. A. Lahankar, and A. G. Suits, Rev. Sci. Instrum. 76, 063106 (2005). doi:  10.1063/1.1921671
    [34] M. Cheng, Z. J. Yu, L. L. Hu, D. Yu, C. W. Dong, Y. K. Du, and Q. H. Zhu, J. Phys. Chem. A 115, 1153 (2011). doi:  10.1021/jp106624q
    [35] L. Rubio-Lago, J. D. Rodríguez, A. García-Vela, M. G. González, G. A. Amaral, and L. Bañares, Phys. Chem. Chem. Phys. 13, 8186 (2011). doi:  10.1039/c0cp02515a
    [36] W. K. Qi, P. Jiang, D. Lin, X. P. Chi, M. Cheng, Y. K. Du, and Q. H. Zhu, Rev. Sci. Instrum. 89, 013101 (2018). doi:  10.1063/1.5006982
  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Figures(11)  / Tables(1)

Article Metrics

Article views(64) PDF downloads(31) Cited by()

Proportional views
Related

A Three-Dimensional Velocity-Map Imaging Setup Designed for Crossed Ion-Molecule Scattering Studies

doi: 10.1063/1674-0068/cjcp2012219

Abstract: In this study, we report the design and simulation of an electrostatic ion lens system consisting of 22 round metal plates. The opening of the extractor plate is covered with metal mesh, which is for shielding the interaction region of the lens system from the high DC voltages applied to all other plates than the repeller and extractor plates. The Simion simulation shows that both velocity-mapping and time focusing can be achieved simultaneously when appropriate voltages are applied to each of the plates. This makes the ion lens system be able to focus large ionic volumes in all three dimensions, which is an essential requirement for crossed ion-molecule scattering studies. A three-dimensional ion velocity measurement system with multi-hit and potential multi-mass capability is built, which consists of a microchannel plate (MCP), a P47 phosphor screen, a CMOS camera, a fast photomultiplier tube (PMT), and a high-speed digitizer. The two velocity components perpendicular to the flight axis are measured by the CMOS camera, and the time-of-flight, from which the velocity component along the flight axis can be deduced, is measured by the PMT. A Labview program is written to combine the two measurements for building the full three-dimensional ion velocity in real time on a frame-by-frame basis. The multi-hit capability comes from the fact that multiple ions from the camera and PMT in the same frame can be correlated with each other based on their various intensities. We demonstrate this by using the photodissociation of CH3I at 304 nm.

These authors contributed equally to this work.
Guo-dong Zhang, Li-chang Guan, Zi-feng Yan, Min Cheng, Hong Gao. A Three-Dimensional Velocity-Map Imaging Setup Designed for Crossed Ion-Molecule Scattering Studies[J]. Chinese Journal of Chemical Physics , 2021, 34(1): 71-80. doi: 10.1063/1674-0068/cjcp2012219
Citation: Guo-dong Zhang, Li-chang Guan, Zi-feng Yan, Min Cheng, Hong Gao. A Three-Dimensional Velocity-Map Imaging Setup Designed for Crossed Ion-Molecule Scattering Studies[J]. Chinese Journal of Chemical Physics , 2021, 34(1): 71-80. doi: 10.1063/1674-0068/cjcp2012219
  • Studying ion-molecule scattering dynamics is relevant to many different research fields, such as interstellar and circumstellar chemistry, planetary atmospheric chemistry, plasma physics, combustion dynamics, and so on [1-3]. There have been many experimental methods developed for investigating ion-molecule scattering dynamics. From early times to nowadays, the flow and drift tubes have been among the most important methods for the acquisition of ion-molecule reaction rate coefficients [4-7]. Precise measurements of integral cross-sections for ion-molecule scatterings can be obtained by using various guided ion beam techniques, and the branching ratios of different reaction channels can also be measured [8-10]. Quantum state selected ion beams have been prepared based on the photoionization method and have been applied to the guided ion beam experiments to study how the integral cross-sections and branching ratios depend on specific quantum states of the ionic reactants [11-13]. Recently, the developments of electric or magnetic based techniques for manipulating molecular beams combined with the Coulomb crystal method have enabled the studies of various effects of ion-molecule reactions, such as the effects of conformation [14], molecular rotation [15], isotope substitution [16], and so on. All these methods focus on obtaining the integral information of the scattering process, little information about the angular distribution of the scattering products in the three-dimensional space can be obtained, which however contains the most detailed scattering dynamics information.

    The crossed molecular beam method combined with the velocity-map imaging (VMI) technique has recently become an important way for studying the ion-molecule scattering dynamics, which measures the three-dimensional angular distributions of the scattering products. Good examples can be found from the research groups of Farrar [17], Wester [18-20] and Tian [21-23]. Compared to the VMI apparatuses used in conventional photodissociation and neutral crossed molecular beam scattering experiments [24-26], the electro-static VMI lens system for ion-molecule scattering experiments has several additional requirements. First, the crossing region of the ion-molecule scattering experiments needs to be field free before and during the scattering process, and right after the reaction is finished, a fast high voltage pulse should be applied to the VMI system to push the charged products to the detector; this requires that the interaction region needs to be well shielded from any high voltage sources. Second, the volume of the interaction region is usually much larger in all three dimensions compared with those in the conventional photodissociation and neutral crossed molecular beam scattering experiments, where a focused laser beam for photoionization is used; this requires that the VMI system should have both velocity-mapping and time focusing simultaneously at relatively large scales. Finally, the signal level for crossed ion-molecule scattering experiments is much lower, usually on the order of several tenths of ions per experimental cycle; this makes the full three-dimensional velocity detection a preferred choice over the conventional crushed two-dimensional or the time-sliced imaging methods. Both the Wester and Tian groups have made special designs for their setups, and different full three-dimensional velocity detection methods have been applied [22, 27, 28].

    In this study, we present the design, simulation and characterization of an electro-static VMI lens system, which implements both velocity-mapping and time focusing simultaneously, and can focus large ionic volumes in all three dimensions, thus suitable for crossed ion-molecule scattering studies. The three-dimensional velocity detection method based on a microchannel plate (MCP), a P47 phosphor screen, a CMOS camera with USB3 connector, a fast photomultiplier tube (PMT) and a high-speed digitizer as first developed by Li and Suits is adopted here [29, 30]. We have written our own Labview program for building the full three-dimensional ion velocity in real time on a frame-by-frame basis. The photodissociation of CH$_3$I at 304 nm has been used for demonstrating the ability of the system.

  • The overall schematic diagram of the crossed ion-molecule scattering setup is shown in FIG. 1, which includes the VMI stack, the three-dimensional velocity detection system and the neutral supersonic molecular beam source generated by a pulsed piezoelectric valve. The pulsed ion source will not be described in this work, thus not shown in the figure. We will describe the VMI stack and three-dimensional velocity detection system in detail in this and next sections, respectively.

    Figure 1.  Schematic diagram of the three-dimensional velocity-map imaging setup and the demonstration of a typical frame which contains two ion events with different intensities on both the camera and PMT.

    To have a large focus volume as required by crossed ion-molecule scattering studies, the VMI stack that we used here consists of 22 stainless steel round plates of 1 mm thickness, similar to the recent designs by Lin et al. [25] and Yu et al. [26]. All the plates have outside diameters of 90 mm except for the first three, which are 104 mm in diameter. The extractor plate (the second plate) has an opening with a diameter of 40 mm, and all other plates from the third to the last have openings with a diameter of 50 mm. The spacing between the repeller (the first plate) and extractor is 15 mm, allowing enough space for reactant ions, neutral molecules and laser beams to pass through; all other plates are separated from each other by 10 mm.

    As this VMI stack is designed specifically for crossed ion-molecule scattering studies, the crossing region between the repeller and extractor needs to be field free during the reaction, and fast high voltage pulses have to be applied to the VMI stack to extract the product ions when the reaction is done. It is obvious that pulsing all the 22 plates at the same time is unpractical in real experiments. Instead, we only pulse the first three plates, and DC voltages are applied to all other plates all the time. In order to shield the crossing region from the high DC voltages of all other plates, the opening of the extractor plate is covered by a thin metal mesh with a transmission efficiency of 88%. We have also made the diameters of the first three plates larger (104 mm in diameter) to avoid the possible high DC voltage leakage into the crossing region through the edges of the plates. A shielding tube with 8 holes (10 mm in diameter) in symmetric positions is used to cover the first five plates as shown in FIG. 1. This tube is not only for better electric shielding of the crossing region, but also crucial for the focusing and deceleration of the reactant ions, which will not be discussed further in the current study.

    One advantage of using multiple plates instead of three as in the conventional VMI setup [24] is that more parameters can be scanned in order to achieve better focus in all three dimensions [25]. We have used the Simion simulation (Simion 8.0.4) to characterize and optimize our VMI stack, and achieved both velocity-mapping and time focusing simultaneously, which is crucial for the three-dimensional velocity detection of the product ions generated from crossed ion-molecule scatterings. The optimized voltages to all the 22 plates are listed in Table Ⅰ, and the electric potential simulation of the VMI stack with ion trajectories is shown in FIG. 2.

    Table Ⅰ.  Optimized voltages (in Ⅴ) applied to the VMI stack.

    First, we characterize the focus ability of the VMI stack on the two dimensions parallel to the MCP detector (the $XY$ plane as shown in FIG. 2). To do this, we put a line source of ions with different sizes along $Y$ axis ($X$ axis is similar due to the cylindrical symmetry of the VMI stack) in the center of the VMI stack, all ions have the same mass of 127 (I$^+$ ion) and the same kinetic energy of 1 eV along $Y$ axis. FIG. 2 shows a typical simulation with an ion source of 10 mm. We measure the size of the focal width on the MCP detector, and plot it with respect to the size of the ion source, as shown in FIG. 3. It can be seen from FIG. 3 that ion source with a length up to about 12 mm can be very well focused through the VMI stack, and a velocity resolution of $\sim$1% at 1 eV can be reached even for an ion source of 22 mm long in an ideal scenario, considering the fact that a 75 mm MCP detector is used in the current setup.

    Figure 2.  Simion simulation of the VMI stack. The ion trajectories shown correspond to ions starting from the middle of the first two plates with 1 eV initial kinetic energy along Y axis. The length of the line shape ion source along Y axis is 10 mm. The distance between the ion source and MCP is designed to be 650 mm, while the real distance might be around 645 mm, see text for details.

    Figure 3.  The size of the focused ion spot on the MCP detector (focal width) through the VMI stack as a function of the initial length of the ion source, see text for details.

    Second, for crossed ion-molecule scattering experiments, the size of the ionic volume along the central axis of the VMI stack ($Z$ axis in FIG. 2) is usually on the order of several millimeters, thus how the performance of the VMI stack differs when the ion source is on and off from the middle between the first two plates along $Z$ axis is crucial. To test this, we move an ion source of 10 mm (mass: 127, kinetic energy: 1 eV along $Y$ axis) around the center of the VMI stack along $Z$ axis, and the size of the focal width on the MCP detector is plotted in FIG. 4. It can be seen that the focal width increases when the ion source is off from the center of the VMI stack, thus the kinetic energy resolution decreases. However, within an offset of $\pm$2 mm, the absolute resolution is still in the range of $\sim$1% in ideal conditions, which is more than enough for our purposes.

    Figure 4.  The size of the focused ion spot on the MCP detector (focal width) through the VMI stack, when moving an ion source of 10 mm (mass: 127, kinetic energy: 1 eV along Y axis) around the center of the VMI stack along Z axis, see text for details.

    The third parameter to investigate is the magnification factor $M_{\rm{s}}$, which is defined in the equation

    where $V_x$ and $V_y$ are the velocities along $X$ and $Y$ axis, respectively, ($x$, $y$) are the coordinates of the ion on the MCP detector, ($x_0$, $y_0$) are the coordinates for ions with zero velocity on the MCP detector, and $t_{\rm{flight}}$ is the time-of-flight of the ion. It is important in our case that the magnification factor $M_{\rm{s}}$ does not depend significantly on the mass of the ion, the absolute voltages applied and the offset along $Z$ axis from the VMI center. We simulate a series of ions with different velocities, and obtain the magnification factor $M_{\rm{s}}$. The simulations do show that $M_{\rm{s}}$ is almost independent of the mass of the ion and the absolute voltages applied, while it does weakly depend on the offset along $Z$ axis from the VMI center, as presented in FIG. 5. The change of $M_{\rm{s}}$ from the offset at +1 mm to that at -1 mm is only $\sim$0.8%, which is negligible under our current experimental conditions. We have collected the time-slice VMI images for the photodissociation of CH$_3$I at 304 nm (the experimental details will be described later) at different offsets between +2 and -2 mm, and no obvious differences on either the resolution or the image sizes can be noticed, which shows that the performance of the VMI stack does not critically depend on the ion source offsets along $Z$ axis from the center under the current experimental conditions.

    Figure 5.  The magnification factor Ms versus the offset of the ion source from the center of the VMI stack along Z axis, here the mass of the ion is 127, see text for details.

    The velocity components along $X$ and $Y$ axes, namely $V_x$ and $V_y$, are measured through the ion position on the MCP detector as described above. The third component $V_z$ is obtained by measuring the time-of-flight of the ion through the equation

    where $M_t$ is the magnification factor along $Z$ axis, and $t_0$ is the time-of-flight when the ion has zero velocity along $Z$ axis. Two requirements need to be fulfilled for the above procedure to work correctly. First, the time-of-flight $t_{\rm{flight}}$ needs to be linearly dependent on $V_z$; second, it requires that the VMI stack needs to have time focusing, that is, ions starting from different initial positions along $Z$ axis with the same velocity component $V_z$ should arrive at the detector at the same time [31]. The first requirement can be easily fulfilled as shown in FIG. 6, where $t_{\rm{flight}}$ is plotted as a function of $V_z$. It can be seen that a perfect linear correlation between $t_{\rm{flight}}$ and $V_z$ exists. As described in Ref.[31], the time focusing condition is solely determined by the voltage ratio on the first two plates, however our simulation shows that voltages on other plates also have slight effect on the time focusing condition of the whole system. We have to adjust all the voltages to reach the time focusing condition, and at the same time keep the velocity-mapping condition on the $X$ and $Y$ dimensions. The simulated result for ions of mass 127 with a velocity of 0 m/s is presented in FIG. 7. It is shown that within an offset in the range of $\pm$1 mm from the VMI center, the $t_{\rm{flight}}$ is almost constant. The time focusing condition has been confirmed experimentally by using Ar$^+$ ions generated through the resonance-enhanced multiphoton ionization (REMPI) method, which will not be presented here.

    Figure 6.  The simulated time-of-flight tflight for ions of mass 127 versus the velocity component along Z axis, Vz.

    Figure 7.  The simulated time-of-flight tflight for ions of mass 127 with velocity of 0 m/s versus the offset from the VMI center along Z axis.

  • There have been many different ways for measuring all the three velocity components of an ion from a single measurement [32]. Recently, Tian group has successfully applied the delay-line detector (DLD) to their crossed ion-molecule scattering setup for three-dimensional velocity detection of the reaction products [22]. Wester group has adopted a conventional two-dimensional velocity detection system, which consists of a MCP detector, a P47 phosphor screen and a CCD camera, combined with a PMT for detecting the time-of-flight of the ion, from which the third velocity component along the axis perpendicular to the detector plane can be deduced [19, 27]. We have adopted the method recently developed by Li, Suits and coworkers [29, 30], which is very similar to that as used by Wester group. The biggest difference between these two methods is that a maximum of only one ion can be detected per experimental cycle in Wester's setup, while multiple ions can be detected in each experimental cycle by using the method of Li and Suits, due to the use of a fast digitizer [29, 30]. This can substantially increase the data collection efficiency of the system.

    The schematic diagram of our three-dimensional velocity detection system is shown in FIG. 1. It consists of a MCP detector (diameter of 75 mm) equipped with a P47 phosphor screen supplied by Photek, a CMOS camera (Model UI-3060CP-M-GL Rev.2 from IDS) with USB3 connection to the computer, which ensures fast enough real-time data transfer rates, a PMT (Hamamatsu CR131) and a fast digitizer (NI PXIe-5160), which can transfer the time-of-flight traces generated by the PMT to the computer for real-time analysis. During each experimental cycle, the CMOS camera captures a two-dimensional image (pixel resolution 968$\times$608) of the phosphor screen, which contains the position information of all the ions; at the same time the PMT and fast digitizer capture the time-of-flight trace, which contains the time of arrival for all the ions. The CMOS camera and the digitizer send the image and time-of-flight trace to the computer in real time. In the computer, a Labview program has been written for real time data analysis. When a frame of data arrives, the image analysis module of the program counts the number of ions captured by the camera and calculates their positions and intensities. The positions are calculated by using the center-of-mass method [33], and the intensities are simply the summation of all the pixels covered by the same ion; at the same time, the time-of-flight analysis module of the program counts the number of peaks detected in the corresponding time-of-flight trace, and calculates the peak amplitudes and time positions of the rising edges. After the above calculations are done, the program compares the number of ions captured by the camera with the number of peaks detected by the PMT, if they are different from each other, this frame of data will be discarded; if they are equal to each other, the program will sort the ions according to their intensities from the camera and peak amplitudes from the PMT. By doing so, we can correlate the multiple ion positions measured by the camera with the ion time-of-flights measured by the digitizer to obtain the full three-dimensional velocities. All these have to be done in real time on a frame by frame basis. A crucial point for the above procedure to work correctly is the synchronization between the camera and the digitizer. Since the camera needs longer time for initial configuration, we let the digitizer wait until the camera is ready, then they start to collect data at the same time. We also monitor the time consumption of the program for every experimental cycle to make sure it can handle all data in real time.

    Compared with the DLD method as used in Tian's group [22], the current method has a longer dead time between adjacent ion hits, which is mainly limited by the decay time of the P47 phosphor screen used. It has a typical decay time of 80-100 ns, thus two ions arriving at the detector within a 80 ns duration may not be distinguished from each other. However, this limitation can be overcome if the time-of-flight trace is obtained from the MCP directly as demonstrated by Li and Suits [30]. The method described in this study requires real-time analysis of the image and time-of-flight trace, which limits the experiment to run at relatively low repetition rates, typically 10-30 Hz if a relatively high image resolution is needed, while the DLD method can easily go into the kHz range [22]. On the other hand, the current method also has several appealing advantages. First, people can switch between conventional two-dimensional ion imaging and three-dimensional imaging as described above freely. This is sometimes convenient when sophisticated three-dimensional imaging is not needed. Second, a high voltage pulse gate can be applied to the detector, so that only ions of interest are detected. This is important for crossed ion-molecule scattering studies, since the intensities of reactant ions are usually 3-5 orders of magnitude higher than those of the product ions.

    The multi-hit and multi-mass detection capability of the current method has been demonstrated by Li and Suits, and a maximum detection of 27 ions per experimental cycle was reached [30]. In our case, the maximum of ion events per experimental cycle that has been reached is in the range of 6-8. As the number of ions per experimental cycle increases, the percentage of false correlation between the data collected from the camera and PMT also increases, this is mainly due to the limited dynamical ranges of intensity detections by the camera and PMT [29]. If two ions produce light spots on the phosphor screen with intensities close to each other in the same experimental cycle, false position and time correlation can occur, as shown by those faint ring structures as observed in FIG. 8 (b), (c), (e), and (f). We tested that when the signal level is about 2-3 ion events per experimental cycle, the percentage of false correlation is negligibly low.

    Figure 8.  (a) The three-dimensional image of I(2P3/2) generated from the photodissociation of CH3I at 304.67 nm, and (b−f) the time-sliced images at different times. The faint rings as observed in (b), (c), (e) and (f), which have diameters as same as that of the ring in (d), are due to false correlations between the data collected from the camera and PMT; the ion signals beyond the pixel number of ~450 in (a) are due to photodissociation of the residual CH3I in the chamber, see text for details.

  • In this section, we demonstrate the three-dimensional velocity detection system by using the photodissociation of CH$_3$I at 304 nm, which has been well studied before [34, 35]. A 10 Hz pulsed supersonic molecular beam of CH$_3$I seeded in Ar with a stagnation pressure of 30 psi is generated by a piezoelectric valve (Amsterdam Valve, ACPV2-150). The beam passes through a skimmer (Beam Dynamics, Model 50.8) with an aperture diameter of 2 mm, which is $\sim$10 cm from the nozzle. The distance between the skimmer aperture and the VMI center is $\sim$10 cm. The UV laser is generated by doubling the fundamental output of a dye laser (LiopTec, LIOPSTAR-E), which is pumped by the second harmonic output of 10 Hz YAG laser (Beamtech, Nimma-900). The UV laser is fixed at 304.67 nm, and focused into the VMI center by a plano-convex lens with a focal length of 300 mm. The focused UV beam dissociates the CH$_3$I molecules, and at the same time photoionizes the product I($^2$P$_{3/2}$) for detection via a 2+1 REMPI scheme [34, 35]. We will measure the velocities of the product I($^2$P$_{3/2}$) by using both the three-dimensional velocity detection method as described above and also the conventional two-dimensional time-sliced VMI method, and the results will be put together for comparison.

    A typical three-dimensional image for the detected I($^2$P$_{3/2}$) fragment generated in the photodissociation process of CH$_3$I at 304.67 nm is shown in FIG. 8(a). Those ion signals beyond the pixel number of $\sim$450 along the $X$ axis are from photodissociation of the residual CH$_3$I in the chamber. The residual CH$_3$I do not have an overall velocity in the lab frame, thus are separated from those ion signals generated from the supersonic molecular beam. We have performed offline slicing on the detected three-dimensional image at different times with a slicing duration of 30 ns, which are shown in FIG. 8(b)-(f). The sliced images are similar to those if a real high voltage pulse gate with a width of 30 ns is applied to the MCP detector, which proves that the three-dimensional image detection system is working correctly. The measured time-of-flight $t_0$ for I$^+$ ions is 17.980 μs, while the simulated value is 18.070 μs. This difference is due to the uncertainty of the real position of the MCP detector. If we assume the distance between the front face of the MCP detector and the VMI center is 645 mm instead of the designed 650 mm, then the simulated time-of-flight $t_0$ for I$^+$ ions becomes 17.989 μs, which is much closer to the measured value. The optimal voltages applied to the VMI stack are found to only weakly depend on the distance between the front face of the MCP detector and the VMI center, only a change of 1-2 V (on a total voltage of 2000 V) is needed when changing from 650 mm to 645 mm. Based on this, we have adopted the value of 645 mm as the real distance between the MCP detector and the VMI center for all our later studies.

    To convert the detected position and time-of-flight information into velocities, we need to obtain the magnification factors $M_{\rm{s}}$ and $M_{\rm{t}}$ either through simulations or direct measurements on known systems. For $M_{\rm{t}}$, we have used the simulated value by using the linear fitting as shown in FIG. 6. For $M_{\rm{s}}$, in principle we can use the simulated value as shown in FIG. 5. However, to do so we need to find the conversion factor between the pixel number (which is the directly measured value by the camera) and the real distance, which could introduce additional uncertainties. To avoid this, we have used the photodissociation of O$_2$ at 224.999 nm to calibrate the velocity detections on the $X$ and $Y$ dimensions as people have done in conventional two-dimensional imaging detection [25, 36]. We use the magnification factors thus obtained to convert the image as shown in FIG. 8(a) into the velocity space, and slice it on the three different central planes with a width of 50 m/s, which are shown in FIG. 9(a)-(c). The conventional sliced image by applying a high voltage pulse gate with a width of 50 ns to the MCP detector is also shown in FIG. 9(d) for comparison. The obtained velocity distributions from the above two methods are presented in FIG. 10. As seen from FIG. 10, the central velocity measured by the two-dimensional time-sliced imaging method is $\sim$510 m/s, and that by the three-dimensional imaging method is $\sim$505 m/s, which is consistent with the value reported in literatures [34, 35]. The difference between the above two values is $\sim$1%. This is mainly due to the uncertainties on the determination of the magnification factor $M_{\rm{s}}$. In our experimental geometry, the molecular beam is parallel to the plane of the MCP detector, and the velocity spread of the molecular beam can blur the two-dimensional time-sliced image of O$_2$ photodissociation at 224.999 nm (in our case, the obtained best velocity resolution $\Delta v/v$ is $\sim$2%), which is used to obtain the magnification factor $M_{\rm{s}}$. At the same time, the two-dimensional time-sliced image obtained is not ideally symmetric along all directions. Both of the above two factors could cause uncertainties when we use the linear fitting method based on the two-dimensional time-sliced image of O$_2$ photodissociation at 224.999 nm to calibrate the magnification factor $M_{\rm{s}}$. Nevertheless, the uncertainty of $\sim$1% is acceptable for our later applications. Thus, our three-dimensional velocity detection system has been correctly calibrated.

    Figure 9.  Central slices with a width of 50 m/s on the (a) X Z, (b) Y Z, and (c) X Y planes for the three-dimensional image as shown in FIG. 8(a); (d) the conventional two-dimensional time-sliced image.

    Figure 10.  The velocity distributions of the I(2P3\2) fragment generated in the photodissociation process of CH3I at 304.67 nm measured by the conventional two-dimensional time-sliced imaging method (black curve) and the threedimensional imaging method described in the present study (red curve).

  • In this study, we have designed and characterized a VMI stack, which implements both velocity-mapping and time focusing capabilities. It can focus a large ionic volume not only on the two dimensions parallel to the detector plane, but also on the dimension along the flight axis. This makes it suitable for future crossed ion-molecule scattering studies. The three-dimensional velocity detection system as developed recently by Li and Suits [29, 30] has also been successfully constructed in our lab, which has been tested by using the photodissociation of CH$_3$I at 304 nm. We are now optimizing the quantum state selected pulsed ion source, which is based on the pulsed field photoionization method as demonstrated recently by Ng and coworkers [12]. The detailed results will be presented in our future reports.

  • This work is supported by the Program for Young Outstanding Scientists of Institute of Chemistry, Chinese Academy of Science, and Beijing National Laboratory for Molecular Sciences. Hong Gao and Li-chang Guan are also supported by the K. C. Wong Education Foundation. We thank Prof. Shan Xi Tian and Dr. Jie Hu from University of Science and Technology of China for helpful discussions.

Reference (36)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return