Ultra-high vacuum conditions are ideal for the study of trapped ions. They offer an almost perturbation-free environment, where ions confined in traps can be studied for extended periods of time—facilitating precision measurements and allowing infrequent events to be observed. However, if one wishes to study processes involving molecular ions, it is important to consider the effect of blackbody radiation (BBR). The vast majority of molecular ions interact with BBR. At 300 K, state selection in trapped molecular ions can be rapidly lost (in a matter of seconds). To address this issue, and to maintain state selectivity in trapped molecular ions, a cryogenic ion trap chamber has been constructed and characterized. At the center of the apparatus is a linear Paul ion trap, where Coulomb crystals can be formed for ion–neutral reaction studies. Optical access is provided, for lasers and for imaging of the crystals, alongside ion optics and a flight tube for recording time-of-flight mass spectra. The ion trap region, encased within two nested temperature stages, reaches temperatures below 9 K. To avoid vibrations from the cryocooler impeding laser cooling or imaging of the ions, vibration-damping elements are explicitly included. These components successfully inhibit the coupling of vibrations from the cold head to the ion trap—confirmed by accelerometer measurements and by the resolution of images recorded at the trap center (at 9 and 295 K). These results confirm that the cryogenic ion trap apparatus meets all requirements for studying ion–neutral reactions under cold, controlled conditions.

When an ion meets a neutral molecule, the collision system can undergo a “metamorphosis”—transforming the reactant pair into a product pair. The link between the reactant and product quantum states can often be modified by a tiny change in the collision energy, in the initial internal energy of the reactants, or by altering the isotopic composition—with small differences in the properties of the colliding reactants leading to different outcomes. In order to understand as much as possible about the fundamental quantum dynamics of a reaction system, it is necessary to start with quantum state-selected reactants, in an environment free from contaminants, and with control over the collision energy.

These conditions can be achieved by confining the ionic reactants in a cryogenic ion trap. Selected atomic ions (including singly charged alkaline earth metal ions) can be efficiently laser cooled, whereby they adopt a regular lattice structure known as a Coulomb crystal.1,2 Ions that are not amenable to laser cooling, including most molecular ions, can be sympathetically cooled by collisions with co-trapped laser-cooled ions. Elastic collisions between the laser-cooled and co-trapped species can be very efficient, with the sympathetically cooled ions frequently reaching translational temperatures comparable to those of the laser-cooled species.3–5 Laser-cooled ions are constantly fluorescing, and this fluorescence can be observed with a camera to provide real-time images of the average positions of laser-cooled ions within the trap. In many cases, the presence of non-laser-cooled (or “dark”) ions can be inferred from the lattice positions of the laser-cooled ions. A key benefit of studying reactions in Coulomb crystals is the ability to watch single ions react, in real time. The setup also allows experimentalists to play an active role in controlling collision events (instead of passively observing the outcome of a collision), for example, by carefully preparing individual ions in selected quantum states with focused laser beams.6–11 

Although there are a number of techniques now available for addressing and preparing ionic reactants in selected quantum states, maintaining this state selectivity can be practically challenging for molecular ions. This is because reactions in Coulomb crystals are typically monitored over the course of several minutes; sympathetic cooling between laser-cooled and co-trapped ions involves only elastic collisions and, therefore, does not affect the internal state distribution. The vast majority of molecular ions (i.e., species other than homonuclear diatomics) interact with the blackbody radiation (BBR) field, and previous studies have found BBR-induced thermalization to occur rapidly (within 10s of seconds) at 295 K.3,12–15 Approaches such as optical pumping and superimposing an ion trap with a magneto-optical trap have been shown to successfully cool the internal degrees of freedom of selected molecular ions.12,13,16–18 Here, we take a more general approach in constructing a cryogenic linear Paul ion trap apparatus, building on the framework first demonstrated with the CryPTEx setup.19,20 The ion trap is surrounded by cryogenic shielding, minimizing BBR-mediated quantum state population redistribution in trapped molecular ions. The shielding also provides the necessary environment for the introduction of cryogenic buffer gases (for buffer gas cooling applications) and allows the chamber surfaces to act as cryo-pumps, reducing the pressure even further. While there are multiple research groups actively working on cryogenic ion trapping using small chip-based, wafer-based, wire, or blade trap designs,21–26 only a small number of “macroscopic” cryogenic linear Paul ion trap setups have been successfully constructed to date. These cryogenic setups have been designed for applications including the study of highly-charged ions and precision measurements of optical clock transitions.19,27–29 It should also be noted that a number of cryogenic multipole (typically 22-pole) traps have been developed for spectroscopic measurements and cold ion–neutral reaction studies.30–32 While 22-pole trap environments offer many advantages, they are not suited for laser cooling or imaging applications (owing to the limited optical access and the shape of the trapping potential). The cryogenic ion trap apparatus introduced in this article will facilitate the study of cold and controlled chemical reactions involving a diverse range of molecular ionic reactants, taking advantage of the properties of Coulomb crystals and employing sensitive detection methods.

In order to study low-temperature reactions between trapped ions and neutral species, a source of cold neutral reactants is needed. Slow-moving and state-selected neutral reactants can be prepared using, for example, the filtered output of a Stark or Zeeman decelerator—with the output of the neutral reactant beam focused into the cryogenic ion trap and any non-target species blocked or deflected away.33–36 In this way, with the ionic reactants effectively stationary, the collision energy is given by the speed of the neutral beam. State-of-the-art deceleration methods can produce beams of neutral species with velocities as low as 10s of m/s, as seen in decelerators interfaced with neutral particle traps.33–35,37–41 Both Stark and Zeeman decelerators produce state-selected beams of slow-moving polar or radical species, respectively, as the time-varying fields address only a certain sub-set of particles (typically those in low-field-seeking quantum states). With the addition of a magnetic guide at the end of a Zeeman decelerator, only the state- and velocity-selected radical species of interest are admitted to the ion trap region.34,36

The ability to mass-analyze the contents of the ion trap at selected reaction times is an essential requirement for detailed reaction kinetics and dynamics studies in complex chemical systems. This is because imaging methods alone provide insufficient details about the identities and relative numbers of product ions in systems where multiple product channels are open.1,2 In order to quantitatively establish the relative numbers of each ionic species present in the crystal at a given time, ions can be extracted into a time-of-flight (TOF) mass spectrometer. TOF methods have been successfully adopted by a number of research groups for studying reactions in Coulomb crystals.42–47 Here, we report the first cryogenic ion trap setup that features an integrated mass spectrometer for chemical reaction studies.

Designing the cryogenic ion trap (as described in this article) represents a significant technical challenge, owing to the long list of limitations and requirements. Not only is line-of-sight access to the trap center needed for laser cooling and time-of-flight ejection—the cold head also has to be accommodated close to the trap. A novel off-axis parabolic (OAP) mirror array-based imaging setup has been implemented, removing the line-of-sight access requirement for the imaging detection of ions. The OAP mirror relay imaging approach has previously been demonstrated to work at 295 K48 and is schematically illustrated in Fig. 1. Here, the success of the OAP mirror-based imaging setup is demonstrated under cryogenic temperatures (at 9 K). The use of a pulse-tube cryocooler also introduces the possibility of vibrations affecting the resolution of Coulomb crystal images. This concern has been recognized in other cryogenic ion trap designs, with innovative methods for minimizing vibrations proposed.27,49 Here, we make several design choices that serve to reduce the effect of vibrations on the innermost trap region. We present the design of the cryogenic ion trap apparatus constructed in our laboratory, describing the novel elements in detail and quantitatively assessing their performance.

FIG. 1.

Cut view of the imaging system. Images from the center of the ion trap (a) are passed to the microscope objective (d) via an optical relay consisting of two off-axis parabolic (OAP) mirrors (b) and (c). The first OAP mirror (b) is positioned beside the ion trap, within the innermost cold shield inside the vacuum chamber. All other components that make up the imaging system are located outside the vacuum chamber. Numbers 1 and 3 refer to connection points on the Heim column from Fig. 4.

FIG. 1.

Cut view of the imaging system. Images from the center of the ion trap (a) are passed to the microscope objective (d) via an optical relay consisting of two off-axis parabolic (OAP) mirrors (b) and (c). The first OAP mirror (b) is positioned beside the ion trap, within the innermost cold shield inside the vacuum chamber. All other components that make up the imaging system are located outside the vacuum chamber. Numbers 1 and 3 refer to connection points on the Heim column from Fig. 4.

Close modal

The ion trap is based on a conventional linear quadrupole radio-frequency trap design, with a rod separation of 7 mm and a rod diameter of 8 mm. Each rod is segmented into three components, with the central segment 5.4 mm long and the end segments each 20 mm long. The segments are separated by 0.2 mm. This geometry is adopted as it allows for the formation of Coulomb crystals containing up to several hundred ions. While implementing a surface trap would be technically easier, such traps cannot be easily employed for reaction studies (or for the confinement of more than a few tens of ions). As the ion trap will be interfaced with a beam of Zeeman-decelerated and guided neutral species, it will be useful to be able to physically move the entire trap a few mm in all directions, in order to achieve optimal alignment between these components. It is also desirable to have minimal separation between the trapping region and the final component of the magnetic guide to maximize the number of neutral reactants that reach the trapped ions.

Given the need to surround the trap with cryogenic shielding, the aforementioned requirements are met by reducing the physical size of the trap itself: the trapping rods are cut in half along the trapping axis and are mounted on a cryogenic shield that encloses the trapping region (see Fig. 2). The rods and the trap housing are machined from oxygen-free, high-conductivity (OFHC, 99.99% purity) copper to ensure maximum thermal conductivity between the cold head and the trapping region. The screws holding the trap in place are electrically insulated from the shielding by washers machined from a ceramic material (Shapal Hi-M Soft, Precision Ceramics) with a favorable thermal expansion coefficient, good thermal conductivity, and with a sufficiently large dielectric constant. The screws (stainless steel) also serve as connection points for the radio-frequency and ion extraction circuitry. The rods themselves are electrically insulated from the shielding by additional thin layers (0.6 mm) of Shapal placed between the rods and the shielding. Finite-element method simulations, carried out to establish if any major distortion of trapping fields are predicted by the trap design and shielding, confirm that no significant distortions are expected to arise from the shielding around the ion trap region. The walls of the shielding surrounding the trap are near-symmetric, with small openings between the central trapping regions of all four rods. These openings serve several purposes: they allow for the admission of neutral Ca atoms (for the formation of Ca+ ions), provide laser access, facilitate Coulomb crystal imaging, and allow for the extraction of ions into a time-of-flight tube for recording mass spectra.

FIG. 2.

Schematic illustration of the linear quadrupole ion trap, with part of the region cut away to make the interior visible. The trap rods are depicted in yellow, with the surrounding ion trap holder in red (with a blue cross section). The rod segments are held in place, and electrical connections are secured, by a series of small orange screws. One of the ceramic insulators (between the rod and the ion trap holder) can be seen on the right-hand side, in gray.

FIG. 2.

Schematic illustration of the linear quadrupole ion trap, with part of the region cut away to make the interior visible. The trap rods are depicted in yellow, with the surrounding ion trap holder in red (with a blue cross section). The rod segments are held in place, and electrical connections are secured, by a series of small orange screws. One of the ceramic insulators (between the rod and the ion trap holder) can be seen on the right-hand side, in gray.

Close modal

The innermost region of the chamber, housing the ion trap, is designed to reach temperatures 10 K. The radiation shielding surrounding the ion trap (see Figs. 2 and 3), in addition to the OAP mirror, is physically connected to the second stage of the cold head. This is housed in an outer radiation shield connected to the first stage of the cold head (getting down to temperatures around 60 K). All of the ion trap wires and sensor wires are anchored to the shielding or the cold head to avoid placing additional heat load on the ion trap region. Due to the close proximity between the OAP mirror and the ion trap, chimneys were included in the inner and outer shielding to minimize the transmission of 300 K BBR into the trap center while maintaining optical access (see Ref. 48 for further details on the design of the imaging system).

FIG. 3.

Schematic illustration of the cryogenic ion trap chamber and imaging system. The ion trap (shown in red) is nested inside several layers of shielding. Both the ion trap (within the ion trap holder) and an OAP mirror are encased within the second cold shield, also shown in red. This second cold shield is connected to the second stage of the cryocooler. The first cold shield, in yellow, is built around the second cold shield and connects to the first stage of the cryocooler. Calcium atoms are generated by the calcium oven, shown in blue on the lower left-hand side of the figure. Fluorescence from the laser-cooled ions and images of the back-lit Ronchi ruling are directed by the first OAP mirror (shown in dark gray and to the immediate right of the ion trap, within the red shielding region) to the second OAP mirror, situated outside the vacuum chamber (held in place by a blue mount), and through the rest of the imaging relay system on the upper (blue) breadboard. The pathway followed by the ionization and laser cooling beams is illustrated by the thin (left) red line passing through the center of the ion trap. The imaging pathway is illustrated by the parallel (right) thin red line, with the images relayed out of the chamber by an OAP mirror. The passage of Ca atoms from the oven to the trap center is illustrated in green. At the trap center, Ca atoms are non-resonantly ionized by a laser (at 355 nm), forming Ca+ ions. Additional reactants can be introduced along the second (off-axis) green pathway illustrated. Slow-moving neutral reactants will be introduced along the yellow pathway, where a Zeeman decelerator and magnetic guide will be connected. All other chamber components, including the cold head (on the top right-hand side of the figure) and the Heim column (bottom central region), are depicted in gray. Note that the TOF detection pathway cannot be seen in this schematic drawing, as it is directed into the page. [An alternative cut-away view of the ion optics for TOF detection is provided in Fig. 10 (top).]

FIG. 3.

Schematic illustration of the cryogenic ion trap chamber and imaging system. The ion trap (shown in red) is nested inside several layers of shielding. Both the ion trap (within the ion trap holder) and an OAP mirror are encased within the second cold shield, also shown in red. This second cold shield is connected to the second stage of the cryocooler. The first cold shield, in yellow, is built around the second cold shield and connects to the first stage of the cryocooler. Calcium atoms are generated by the calcium oven, shown in blue on the lower left-hand side of the figure. Fluorescence from the laser-cooled ions and images of the back-lit Ronchi ruling are directed by the first OAP mirror (shown in dark gray and to the immediate right of the ion trap, within the red shielding region) to the second OAP mirror, situated outside the vacuum chamber (held in place by a blue mount), and through the rest of the imaging relay system on the upper (blue) breadboard. The pathway followed by the ionization and laser cooling beams is illustrated by the thin (left) red line passing through the center of the ion trap. The imaging pathway is illustrated by the parallel (right) thin red line, with the images relayed out of the chamber by an OAP mirror. The passage of Ca atoms from the oven to the trap center is illustrated in green. At the trap center, Ca atoms are non-resonantly ionized by a laser (at 355 nm), forming Ca+ ions. Additional reactants can be introduced along the second (off-axis) green pathway illustrated. Slow-moving neutral reactants will be introduced along the yellow pathway, where a Zeeman decelerator and magnetic guide will be connected. All other chamber components, including the cold head (on the top right-hand side of the figure) and the Heim column (bottom central region), are depicted in gray. Note that the TOF detection pathway cannot be seen in this schematic drawing, as it is directed into the page. [An alternative cut-away view of the ion optics for TOF detection is provided in Fig. 10 (top).]

Close modal

In order to reach temperatures 10 K in the ion trap region, a pulse-tube cryocooler is required. The cryocooler utilized in this setup (4 K pulse-tube cryocooler from Sumitomo Heavy Industries, Ltd.) has the capacity to cool down to temperatures of 45 K at the first stage and 4.2 K on the second stage. As the name suggests, the cold head is pulsed; the cryocooler compressor unit operates at a frequency of ∼50 Hz, with the two stages of the cold head vibrating at a frequency of 1–2 Hz.49–51 The vibrations associated with the cryocooler have been carefully considered in the design of the apparatus. The apparatus is made up of two parts, with each of these components sitting on a different frame. One frame supports the outside chamber, the cold head, and part of the first stage cold shield. The second frame—which is nested inside the first frame and mechanically connected to it by a flexible stainless-steel bellow—supports a hexapod and Heim column52 setup, on which the cold shields are mounted. The Heim column serves to isolate the ion trap thermally (from the chamber walls), with the bellow positioned below the Heim column isolating the ion trap vibrationally (from the cold head). The hexapod serves to move the Heim column (and the trap) with respect to the main vacuum chamber, for alignment purposes. The Heim column is made of two titanium tubes and one aluminum tube, constructed such that any thermal shrinkage (when the chamber is cooled down) is compensated for—mitigating the need for extensive optical realignment whenever the chamber is cooled down or warmed up. The design of the Heim column can be seen in Fig. 4.

FIG. 4.

Cut views of the Heim column, with contraction directions indicated. The titanium first stage (1) is connected to the hexapod at room temperature via a cylindrical holder (detail B). The second stage (2) made from aluminum is connected to the first radiation shield of the chamber (reaching temperatures down to ∼60 K). The innermost region of the chamber, containing the ion trap and cooled to ∼10 K, rests on the third stage (3) of the Heim column (detail A). Note that individual stages are attached to each other only on one side and there is a vacuum gap between the walls. Thermal contraction directions of the stages are indicated by (color-coded) arrows (not to scale). The hexapod is attached to a counterweight of 213.6 kg (not shown), providing stability to the column and anchoring it to the ground. A certain degree of vibrational decoupling from the rest of the apparatus (subjected to further analysis later in the text) is achieved by a flexible stainless-steel bellow (detail B), the top flange of which is connected to the vacuum chamber accommodating the cryo-head.

FIG. 4.

Cut views of the Heim column, with contraction directions indicated. The titanium first stage (1) is connected to the hexapod at room temperature via a cylindrical holder (detail B). The second stage (2) made from aluminum is connected to the first radiation shield of the chamber (reaching temperatures down to ∼60 K). The innermost region of the chamber, containing the ion trap and cooled to ∼10 K, rests on the third stage (3) of the Heim column (detail A). Note that individual stages are attached to each other only on one side and there is a vacuum gap between the walls. Thermal contraction directions of the stages are indicated by (color-coded) arrows (not to scale). The hexapod is attached to a counterweight of 213.6 kg (not shown), providing stability to the column and anchoring it to the ground. A certain degree of vibrational decoupling from the rest of the apparatus (subjected to further analysis later in the text) is achieved by a flexible stainless-steel bellow (detail B), the top flange of which is connected to the vacuum chamber accommodating the cryo-head.

Close modal

The first thermal shield is mounted on the outermost section of the Heim column, with the second thermal shield resting on top of the innermost Heim column section. Each of these sections is thermally anchored by flexible copper braids connected to the corresponding stage on the cold head. Previous work has found the propagation of vibrations through these copper braids to be negligible.53 Finally, the valve unit of the cryocooler—an element that needs to be physically close to the cold head—is located on a shelf suspended from the laboratory ceiling. This removes the need for the unit to be in contact with either frame, while keeping it in the vicinity of the cold head.

To verify the ability of the apparatus to reach temperatures 10 K in the ion trap region, temperature sensors have been placed at key positions on the setup. Thermodiodes capable of recording temperatures down to 4 K are attached directly to the cold head and onto the OAP mirror mounting. The ion trap is directly connected to the mirror mounting (both are within the innermost temperature stage) to which the sensor is attached. The temperature has also been monitored at several other places, including on the outer shielding. As can be seen in Fig. 5, the temperature of both the first and second stages goes down very quickly once the cryocooler is turned on. The temperature of the innermost (second) stage can be seen to plateau at 8.89 K, achieving the goal of reaching temperatures below 10 K within 8 h. Our design contains much less bulk material than, for example, the CryPTEx setup,19 so the cooling period is shortened accordingly. The cooling speed is also affected by the thermal connection between the first and third stages of the Heim column, which acts as an imperfect passive thermal switch: at high temperatures, the connection speeds up the cooling process; as the temperature falls, the titanium thermal conductivity is gradually reduced, lessening the influence of the connection between the Heim column stages. The relatively large outer surface of the second stage of the Heim column (∼500 cm2) is in close proximity to the warm first stage (separated by only 4 mm), placing a high thermal load (∼24 W) on the first radiation shield when compared to the maximum cooling power of the first stage of the cold head (∼35 W). As anticipated, the cryocooler can tolerate this thermal load, with the temperature of the first cooling stage plateauing at 61 K. These findings validate the appropriateness of the design choices outlined in Sec. II, confirming that temperatures below 10 K can be attained in the ion trap region.

FIG. 5.

Temperatures recorded at the first (outer radiation shield; dark red) and second (inner radiation shield; orange) stages, plotted as a function of time. The first vertical dashed line indicates when the cryocooler was turned on, and the second dashed line corresponds to when the cryocooler was turned off.

FIG. 5.

Temperatures recorded at the first (outer radiation shield; dark red) and second (inner radiation shield; orange) stages, plotted as a function of time. The first vertical dashed line indicates when the cryocooler was turned on, and the second dashed line corresponds to when the cryocooler was turned off.

Close modal

To establish whether operating at cryogenic temperatures causes any problems for the imaging system, a Ronchi ruling placed inside the trap has been imaged under different conditions. A UV LED is situated behind the Ronchi ruling to provide back-lighting. An OAP mirror positioned beside the trap (inside the chamber) relays the image of the grid to a partner OAP located outside the vacuum chamber, where it passes through a series of optical elements and into an electron multiplying (EM)-CCD camera (see Fig. 1). This novel imaging setup has been designed to overcome the space and optical access limitations imposed by the shielding. As set out in a previous publication, the OAP mirror-based imaging relay was initially tested and characterized outside the vacuum system.48 Here, we demonstrate that it also successfully relays images recorded at the center of the ion trap at both 9 K and at room temperature. This represents an important finding, as the first OAP mirror is located within the innermost nested temperature stage (alongside the ion trap) and is cooled down to 9 K. The performance of the OAP mirror is not hindered by the cryogenic environment.

While the position of the first OAP mirror is fixed when the chamber is under vacuum, all external optical components can be adjusted to ensure that the pair of OAPs are aligned with respect to one another and to optimize image resolution. As can be seen in Fig. 6, the performance of the imaging setup is comparable at room temperature and at 9 K. The resolution is also comparable to that obtained from setting up the components outside the vacuum chamber (on an optical table).48 The features of the ruler can be resolved in both images in Fig. 6, with a scale bar provided for context. In Coulomb crystals, neighboring ions are typically separated by 10–20 μm, corresponding to the spacing between one or two lines on the Ronchi ruler. These findings confirm that the resolution of the OAP mirror-based imaging system is more than sufficient to record sharp Coulomb crystal images, both at room temperature and when the ion trap region is cooled down to 9 K. There are neither obvious low-temperature deformations nor any distortions introduced by vibrations, further validating the design choices outlined above. The Ronchi ruling has also been imaged with the grid lines horizontal (in addition to the vertical alignment shown in Fig. 6), with the same result: excellent resolution is achieved both at room temperature and at 9 K.

FIG. 6.

Images of a Ronchi ruling (100 lines per mm), positioned at the center of the ion trap region, recorded with an EM-CCD camera. The Ronchi ruling is back-lit with a UV LED, with the image relayed by a pair of OAP mirrors to the camera (see Fig. 1). (Left) Recorded with the ion trap region at room temperature (∼295 K), with the cryocooler off. (Right) Taken with the cryocooler on, with the OAP mirror and ion trap at 9 K.

FIG. 6.

Images of a Ronchi ruling (100 lines per mm), positioned at the center of the ion trap region, recorded with an EM-CCD camera. The Ronchi ruling is back-lit with a UV LED, with the image relayed by a pair of OAP mirrors to the camera (see Fig. 1). (Left) Recorded with the ion trap region at room temperature (∼295 K), with the cryocooler off. (Right) Taken with the cryocooler on, with the OAP mirror and ion trap at 9 K.

Close modal

It has been previously documented that pulse-tube cryocoolers generate vibrations—particularly at the pulse tube frequency and at higher harmonics of this frequency. The amplitudes of the vibrations are reported to be 10μm at each of the stages of the cryocooler cold head.49 As this level of vibration is comparable to the inter-ion spacing in Coulomb crystals, it is necessary to minimize vibrations at the position of the ion trap; else, the image resolution will be impeded. To ascertain the extent of the vibrations in this setup, accelerometers are placed at different positions on the cryogenic ion trap chamber and on the two frames supporting the apparatus. In this way, both the frequency and the amplitude of vibrations can be established at different points (and under different conditions) as a function of time.

The displacement of the accelerometer, calculated by double integration of the acceleration over a moving time window of 0.1 ms can be seen in Fig. 7. We estimate the uncertainty in the displacement values to be not more than 30 nm based on the accelerometer broadband resolution (∼5 × 10−4 m/s2 rms), its sensitivity (1000 mV/g), and the accuracy of the analog-to-digital converter. For these measurements, the accelerometer is positioned on the frame that supports the ion trap chamber (partially vibrationally decoupled by a stainless-steel bellow from the main vacuum chamber supporting the turbomolecular pump and the cold head). Interestingly, there is very little difference in the amplitude of the vibrations recorded with the cryocooler on and off. In both cases, the frame is displaced by ∼±500 nm due to vibrations—more than an order of magnitude lower than the amplitude of the vibrations at the cold head quoted by the manufacturer. Accelerometer measurements recorded at other locations on the apparatus, positioned on both the vertical and horizontal axes, achieve comparable findings. The amplitude of the vibrations is not sensitive to orientation or position.

FIG. 7.

Vibrations recorded outside the chamber with an accelerometer situated on the frame supporting the ion trap. The measurements show vibrations in the horizontal plane, recorded with the cryocooler on (black) and off (yellow).

FIG. 7.

Vibrations recorded outside the chamber with an accelerometer situated on the frame supporting the ion trap. The measurements show vibrations in the horizontal plane, recorded with the cryocooler on (black) and off (yellow).

Close modal

In Fig. 8, the frequencies of the vibrations recorded using an accelerometer (with a sensitivity of 500 mV/g) at the ion trap are plotted. Again, it can be seen that there are no major differences between the vibrations measured with the cryocooler on and off, aside from some changes in the relative intensity of the peak at ∼50 Hz (the frequency at which the cryocooler compressor unit, and several other pieces of equipment in the lab, operate at).50Figures 8 and 9—showing the vibrations recorded with the cryocooler on and off, and with the vacuum pumps on and off—clearly illustrate that almost all the vibrations, including the peak at 50 Hz, arise from sources other than the cryocooler. Very pleasingly, there are no obvious features observed below 10 Hz, despite the drive frequency of vibrations at the two stages of the cold head occurring in this region (at 1–2 Hz; the lower bandwith detection limit for both the accelerometers is 0.5 Hz).49,51 Indeed, the absence of any significant low-frequency features in Fig. 8 indicates that vibrations from the cold head are very effectively damped by the apparatus.

FIG. 8.

The frequencies of vibrations recorded with the accelerometer positioned on the ion trap are plotted, recorded both with (black) and without (yellow) the cryocooler on. Measurements are recorded at a sampling frequency of 10 000 Hz. The main plot displays the frequencies and relative intensities of all vibrations, with the majority of these arising from sources other than the cryocooler. The inset (right corner) is a zoomed-in plot of the low-frequency region.

FIG. 8.

The frequencies of vibrations recorded with the accelerometer positioned on the ion trap are plotted, recorded both with (black) and without (yellow) the cryocooler on. Measurements are recorded at a sampling frequency of 10 000 Hz. The main plot displays the frequencies and relative intensities of all vibrations, with the majority of these arising from sources other than the cryocooler. The inset (right corner) is a zoomed-in plot of the low-frequency region.

Close modal
FIG. 9.

Frequencies of vibrations recorded when the cryocooler is off, with the vacuum pumps on (black) and off (yellow). Large differences between the “on” and “off” traces can be seen, especially in the peak at 50 Hz. The inset (right corner) is a zoomed-in plot of the low-frequency region.

FIG. 9.

Frequencies of vibrations recorded when the cryocooler is off, with the vacuum pumps on (black) and off (yellow). Large differences between the “on” and “off” traces can be seen, especially in the peak at 50 Hz. The inset (right corner) is a zoomed-in plot of the low-frequency region.

Close modal

Several high-frequency vibrations are observed due to the presence of other sources of vibration in the laboratory, such as turbomolecular vacuum pumps (which operate at a frequency of 1000 Hz). It is not our intention to compensate for all possible sources of vibration; we set out to reduce the amplitude of the vibrations to <1μm at the position of the ion trap. This goal has been achieved with the design detailed in this manuscript—confirmed by the accelerometer measurements and by the resolution of images recorded at the trap center (with the cryocooler on and off). By making minor amendments to the way that the vacuum pumps are connected, we anticipate that vibrations arising from the pumps will be further suppressed. This is ongoing work, and we hope to reduce the magnitude of vibrations by another factor of 10.

In another cryogenic ion trap system, constructed for the study of highly charged ions and precision spectroscopic measurements, it was essential that the propagation of vibrations between the cryocooler and the ion trap region be minimized. In this apparatus, a 1.4 m separation was introduced between the first and second stages of the cryocooler and the ion trap region—with the cold head and vacuum pumps physically located in another room in the laboratory housing the ion trap chamber. The components were connected by a long arm spanning the two adjoining laboratories, containing “vibration-decoupling” components joined by flexible copper links, with several other vibration-reducing design elements introduced (see Ref. 49 for further details). This very pro-active vibration-damping approach ultimately yielded vibrations on the order of 10 nm at the position of the ion trap—some three orders of magnitude lower than the amplitude of vibrations at the cold head.

For our reaction studies, we do not have the same need for ultra-low vibrations at the position of the ion trap; our requirements are much more modest. Our primary concerns are the ability to obtain sharp images of the Coulomb crystals and to ensure that the alignment of the various lasers (and the imaging system) is not compromised when the apparatus is operated at 9 or 295 K. Thanks to the careful design of the apparatus, with the use of nested frames and the inclusion of a Heim column (see Subsection II B), vibrations have been successfully damped at the position of the ion trap. This is quantified by the accelerometer measurements and verified by the recording of sharp Ronchi ruling images obtained with the cryocooler on and off (as the cryocooler is the primary source of vibrations in the apparatus). Indeed, there is no obvious difference in the resolution of the Ronchi ruling images obtained at 295 or at 9 K (see Fig. 6), verifying that vibrations from the cryocooler do not impede the resolution of the images. These characterization studies also confirm that the resolution of the imaging setup is more than sufficient to distinguish individual fluorescing ions within a Coulomb crystal, at room temperature and under cryogenic conditions.

Sections III A–III C have focused on the attainment of cryogenic conditions, characterization of the imaging setup, and confirmation that vibrations from the cryocooler are suppressed. A number of other design elements have been incorporated into the apparatus to facilitate the study of ion–neutral reactions, as set out in this subsection. In order to form Ca+ Coulomb crystals in the ion trap, a neutral calcium beam source (oven) must be located nearby. To minimize the heat load placed on the cryocooler, the oven is positioned outside the shielded region, in a water-cooled enclosure with a manually retractable shutter (see Fig. 3).

In order to study reactions between ions and neutral species, it is necessary to attach a neutral beam source to the trap (as set out in the Introduction). There are two such sources in our laboratory, and these neutral beam sources are located on frames at slightly different heights. To make the cryogenic ion trap chamber as versatile as possible, the vertical position (height) of the trap chamber is designed to be adjustable within ±10 cm by a motorized stage. If one chooses to perform such a translation, the height of the hexapod must also be altered using a spacer or by adjusting the screws on the six support legs. The latter can also adjust the tilt and the height of the Heim column to ensure that the trap center is aligned with respect to the neutral beam.

Finally, our intention is to use the cryogenic ion trap apparatus to study a diverse range of ion–neutral reaction systems. For quantitative product detection, in addition to the measurement of branching ratios and rate coefficients, TOF mass spectrometry is a very powerful detection method. A number of research groups, including our own, have studied ion–molecule reaction dynamics in Coulomb crystals by converting the trap electrodes into repeller and extractor electrodes at a selected reaction time.33,42–47 This is achieved by switching off the radio frequency trapping fields and applying static ejection fields, causing the contents of the ion trap to be accelerated into a flight tube and onto microchannel plates for detection. Ion optics are explicitly included in the design of the chamber, as can be seen in Fig. 10 (top), to facilitate the recording of TOF mass spectra and ion–neutral reaction studies.

FIG. 10.

(Top) Cut view of the innermost temperature stage of the chamber, showing the ion trap (labeled T) and several ion optics. Ions ejected from the trap are focused and accelerated by lenses 1–9 before entering the flight tube (not depicted) and reaching a microchannel plate (MCP) detector. The aperture of lens 1 is minimized to block BBR from entering the ion trap. (Bottom) Experimental time-of-flight trace, demonstrating that Ca ions can be formed in—and subsequently ejected from—the trap, enabling mass-sensitive analysis of the trap contents. Repeller and extractor voltages are applied to the ion trap rods,43,44 converting the trap into a two-stage Wiley–McLaren style time-of-flight mass spectrometer.54 (For this characterization measurement, lenses 5–9 have been removed and lenses 1–4 are grounded.) The ejected ions are directed through a field-free flight tube and onto a stack of MCPs for detection.

FIG. 10.

(Top) Cut view of the innermost temperature stage of the chamber, showing the ion trap (labeled T) and several ion optics. Ions ejected from the trap are focused and accelerated by lenses 1–9 before entering the flight tube (not depicted) and reaching a microchannel plate (MCP) detector. The aperture of lens 1 is minimized to block BBR from entering the ion trap. (Bottom) Experimental time-of-flight trace, demonstrating that Ca ions can be formed in—and subsequently ejected from—the trap, enabling mass-sensitive analysis of the trap contents. Repeller and extractor voltages are applied to the ion trap rods,43,44 converting the trap into a two-stage Wiley–McLaren style time-of-flight mass spectrometer.54 (For this characterization measurement, lenses 5–9 have been removed and lenses 1–4 are grounded.) The ejected ions are directed through a field-free flight tube and onto a stack of MCPs for detection.

Close modal

In this manuscript, an innovative cryogenic ion trap apparatus featuring a linear Paul ion trap is introduced and characterized. The apparatus is specifically designed for the study of ion–neutral reactions under cold and controlled conditions. There are very few cryogenic ion trap setups featuring “macroscopic” quadrupole Paul traps; most of the existing cryogenic ion trap chambers are built around chip-based or 22-pole traps. While chip-based and multipole trap designs are ideally suited for certain applications, they are not appropriate for the study of reaction dynamics in Coulomb crystals. We identified four key requirements for the study of complex ion–neutral reactions in Coulomb crystals under cold conditions:

  • The ion trap region needs to reach cryogenic temperatures (ideally below 10 K).

  • There must be optical access to the trap center for lasers and to record crystal images.

  • There must be provision for a beam of neutral species to be focused onto the trap center, in order for reactions to occur.

  • A flight tube needs to be included for the recording of TOF mass spectra at selected reaction times.

  • Vibrations need to be minimized to ensure that the alignment of optical components is not compromised and to facilitate the recording of high-resolution crystal images.

The apparatus described here is the first cryogenic ion trap setup (to our knowledge) to include all of these features. While two similar cryogenic ion trap systems—designed for the spectroscopic study of highly charged ions—meet most of these criteria, they lack the ability to quantitatively mass-analyze the trap contents.19,27 A very recently developed cryogenic ion trap featuring a superconducting radio-frequency cavity combined with a linear Paul trap has demonstrated that the trap can also operate as a quadrupole mass filter.29 However, as this apparatus was designed for frequency metrology measurements with highly-charged ions, it lacks the ability to introduce a neutral reactant beam into the trap region. In the cryogenic ion trap system described in this work, several innovative components are introduced to meet these requirements: an OAP mirror-based imaging relay system is adopted, alongside the use of a Heim column and dual-frame structure to decouple the trap from sources of vibration. Characterization measurements confirm that the innermost temperature stage, containing the ion trap, cools down to temperatures below 9 K. Vibrations from the cryocooler are very effectively damped, enabling high-resolution images to be recorded by the OAP mirror-based relay system (at both 9 K and room temperature). While our intended application is the study of cold ion–neutral reactions, we expect that many of these innovations and findings will also be of general interest to the scientific community.

M.H. acknowledges the support from the Primus Research Programme (Project No. PRIMUS/21/SCI/005) of Charles University. B.R.H. is grateful to the Engineering and Physical Sciences Research Council (EPSRC, Project No. EP/N032950/2), the European Commission (ERC Starting Grant, Project No. 948373), and the Community for Analytical and Measurement Science (CAMS fellowship) for funding.

The authors have no conflicts of interest to disclose.

The data that support the findings of this study are openly available in the Liverpool Research Data Catalogue, DataCat.

1.
B. R.
Heazlewood
,
Mol. Phys.
117
(
14
),
1934
(
2019
).
2.
B. R.
Heazlewood
and
H. J.
Lewandowski
,
Emerging Trends in Chemical Applications of Lasers
(
ACS
,
2021
), pp.
389
410
.
3.
I. S.
Vogelius
,
L. B.
Madsen
, and
M.
Drewsen
,
Phys. Rev. Lett.
89
(
17
),
173003
(
2002
).
4.
D.
Zhang
and
S.
Willitsch
,
Cold Chemistry: Molecular Scattering and Reactivity Near Absolute Zero
(
The Royal Society of Chemistry
,
2018
), pp.
496
536
.
5.
X.
Tong
,
A. H.
Winney
, and
S.
Willitsch
,
Phys. Rev. Lett.
105
(
14
),
143001
(
2010
).
6.
P. O.
Schmidt
,
T.
Rosenband
,
C.
Langer
,
W. M.
Itano
,
J. C.
Bergquist
, and
D. J.
Wineland
,
Science
309
(
5735
),
749
(
2005
).
7.
F.
Wolf
,
Y.
Wan
,
J. C.
Heip
,
F.
Gebert
,
C.
Shi
, and
P. O.
Schmidt
,
Nature
530
(
7591
),
457
(
2016
).
8.
C.-w.
Chou
,
C.
Kurz
,
D. B.
Hume
,
P. N.
Plessow
,
D. R.
Leibrandt
, and
D.
Leibfried
,
Nature
545
(
7653
),
203
(
2017
).
9.
M.
Sinhal
,
Z.
Meir
,
K.
Najafian
,
G.
Hegi
, and
S.
Willitsch
,
Science
367
(
6483
),
1213
(
2020
).
10.
Y.
Lin
,
D. R.
Leibrandt
,
D.
Leibfried
, and
C.-w.
Chou
,
Nature
581
(
7808
),
273
(
2020
).
11.
K.
Najafian
,
Z.
Meir
,
M.
Sinhal
, and
S.
Willitsch
,
Nat. Commun.
11
(
1
),
4470
(
2020
).
12.
T.
Schneider
,
B.
Roth
,
H.
Duncker
,
I.
Ernsting
, and
S.
Schiller
,
Nat. Phys.
6
(
4
),
275
(
2010
).
13.
P. F.
Staanum
,
K.
Højbjerre
,
P. S.
Skyt
,
A. K.
Hansen
, and
M.
Drewsen
,
Nat. Phys.
6
(
4
),
271
(
2010
).
14.
N.
Deb
,
B. R.
Heazlewood
,
M. T.
Bell
, and
T. P.
Softley
,
Phys. Chem. Chem. Phys.
15
(
34
),
14270
(
2013
).
15.
N.
Deb
,
B. R.
Heazlewood
,
C. J.
Rennick
, and
T. P.
Softley
,
J. Chem. Phys.
140
(
16
),
164314
(
2014
).
16.
E. R.
Hudson
,
Phys. Rev. A
79
(
3
),
032716
(
2009
).
17.
W. G.
Rellergert
,
S. T.
Sullivan
,
S. J.
Schowalter
,
S.
Kotochigova
,
K.
Chen
, and
E. R.
Hudson
,
Nature
495
(
7442
),
490
(
2013
).
18.
E. R.
Hudson
,
EPJ Tech. Instrum.
3
,
8
(
2016
).
19.
M.
Schwarz
,
O. O.
Versolato
,
A.
Windberger
,
F. R.
Brunner
,
T.
Ballance
,
S. N.
Eberle
,
J.
Ullrich
,
P. O.
Schmidt
,
A. K.
Hansen
,
A. D.
Gingell
 et al,
Rev. Sci. Instrum.
83
(
8
),
083115
(
2012
).
20.
A. K.
Hansen
,
O. O.
Versolato
,
Ł.
Kłosowski
,
S. B.
Kristensen
,
A.
Gingell
,
M.
Schwarz
,
A.
Windberger
,
J.
Ullrich
,
J. R. C.
López-Urrutia
, and
M.
Drewsen
,
Nature
508
(
7494
),
76
(
2014
).
21.
G.
Pagano
,
P. W.
Hess
,
H. B.
Kaplan
,
W. L.
Tan
,
P.
Richerme
,
P.
Becker
,
A.
Kyprianidis
,
J.
Zhang
,
E.
Birckelbaw
,
M. R.
Hernandez
,
Y.
Wu
, and
C.
Monroe
,
Quantum Sci. Technol.
4
(
1
),
014004
(
2018
).
22.
M. F.
Brandl
,
M. W.
van Mourik
,
L.
Postler
,
A.
Nolf
,
K.
Lakhmanskiy
,
R. R.
Paiva
,
S.
Möller
,
N.
Daniilidis
,
H.
Häffner
,
V.
Kaushal
,
T.
Ruster
,
C.
Warschburger
,
H.
Kaufmann
,
U. G.
Poschinger
,
F.
Schmidt-Kaler
,
P.
Schindler
,
T.
Monz
, and
R.
Blatt
,
Rev. Sci. Instrum.
87
(
11
),
113103
(
2016
).
23.
M.
Niedermayr
,
K.
Lakhmanskiy
,
M.
Kumph
,
S.
Partel
,
J.
Edlinger
,
M.
Brownnutt
, and
R.
Blatt
,
New J. Phys.
16
(
11
),
113068
(
2014
).
24.
K.
Geistlinger
,
M.
Fischer
,
S.
Spieler
,
L.
Remmers
,
F.
Duensing
,
F.
Dahlmann
,
E.
Endres
, and
R.
Wester
,
Rev. Sci. Instrum.
92
(
2
),
023204
(
2021
).
25.
C.
Decaroli
,
R.
Matt
,
R.
Oswald
,
C.
Axline
,
M.
Ernzer
,
J.
Flannery
,
S.
Ragg
, and
J. P.
Home
,
Quantum Sci. Technol.
6
(
4
),
044001
(
2021
).
26.
S. L.
Todaro
,
V. B.
Verma
,
K. C.
McCormick
,
D. T. C.
Allcock
,
R. P.
Mirin
,
D. J.
Wineland
,
S. W.
Nam
,
A. C.
Wilson
,
D.
Leibfried
, and
D. H.
Slichter
,
Phys. Rev. Lett.
126
,
010501
(
2021
).
27.
T.
Leopold
,
S. A.
King
,
P.
Micke
,
A.
Bautista-Salvador
,
J. C.
Heip
,
C.
Ospelkaus
,
J. R.
Crespo López-Urrutia
, and
P. O.
Schmidt
,
Rev. Sci. Instrum.
90
(
7
),
073201
(
2019
).
28.
R. J.
Rafac
,
B. C.
Young
,
J. A.
Beall
,
W. M.
Itano
,
D. J.
Wineland
, and
J. C.
Bergquist
,
Phys. Rev. Lett.
85
(
12
),
2462
(
2000
).
29.
J.
Stark
,
C.
Warnecke
,
S.
Bogen
,
S.
Chen
,
E. A.
Dijck
,
S.
Kühn
,
M. K.
Rosner
,
A.
Graf
,
J.
Nauta
,
J.-H.
Oelmann
,
L.
Schmöger
,
M.
Schwarz
,
D.
Liebert
,
L. J.
Spieß
,
S. A.
King
,
T.
Leopold
,
P.
Micke
,
P. O.
Schmidt
,
T.
Pfeifer
, and
J. R.
Crespo López-Urrutia
,
Rev. Sci. Instrum.
92
(
8
),
083203
(
2021
).
30.
D.
Gerlich
,
Phys. Scr.
1995
(
T59
),
256
.
31.
R.
Wester
,
J. Phys. B: At., Mol. Opt. Phys.
42
(
15
),
154001
(
2009
).
32.
J.
Jašík
,
J.
Žabka
,
J.
Roithová
, and
D.
Gerlich
,
Int. J. Mass Spectrom.
354–355
,
204
(
2013
), part of the Special Issue: Detlef Schröder Memorial Issue.
33.
J.
Toscano
,
H. J.
Lewandowski
, and
B. R.
Heazlewood
,
Phys. Chem. Chem. Phys.
22
(
17
),
9180
(
2020
).
34.
B. R.
Heazlewood
,
Annu. Rev. Phys. Chem.
72
,
353
(
2021
).
35.
B. R.
Heazlewood
and
T. P.
Softley
,
Nat. Rev. Chem.
5
(
2
),
125
(
2021
).
36.
J.
Toscano
,
C. J.
Rennick
,
T. P.
Softley
, and
B. R.
Heazlewood
,
J. Chem. Phys.
149
(
17
),
174201
(
2018
).
37.
B. C.
Sawyer
,
B. K.
Stuhl
,
D.
Wang
,
M.
Yeo
, and
J.
Ye
,
Phys. Rev. Lett.
101
(
20
),
203203
(
2008
).
38.
L. P.
Parazzoli
,
N. J.
Fitch
,
P. S.
Żuchowski
,
J. M.
Hutson
, and
H. J.
Lewandowski
,
Phys. Rev. Lett.
106
(
19
),
193201
(
2011
).
39.
Y.
Liu
,
M.
Vashishta
,
P.
Djuricanin
,
S.
Zhou
,
W.
Zhong
,
T.
Mittertreiner
,
D.
Carty
, and
T.
Momose
,
Phys. Rev. Lett.
118
(
9
),
093201
(
2017
).
40.
Y.
Segev
,
M.
Pitzer
,
M.
Karpov
,
N.
Akerman
,
J.
Narevicius
, and
E.
Narevicius
,
Nature
572
(
7768
),
189
(
2019
).
41.
D.
Haas
,
C.
von Planta
,
T.
Kierspel
,
D.
Zhang
, and
S.
Willitsch
,
Commun. Phys.
2
(
1
),
101
(
2019
).
42.
C.
Schneider
,
S. J.
Schowalter
,
K.
Chen
,
S. T.
Sullivan
, and
E. R.
Hudson
,
Phys. Rev. Appl.
2
(
3
),
034013
(
2014
).
43.
N.
Deb
,
L. L.
Pollum
,
A. D.
Smith
,
M.
Keller
,
C. J.
Rennick
,
B. R.
Heazlewood
, and
T. P.
Softley
,
Phys. Rev. A
91
(
3
),
033408
(
2015
).
44.
K. A. E.
Meyer
,
L. L.
Pollum
,
L. S.
Petralia
,
A.
Tauschinsky
,
C. J.
Rennick
,
T. P.
Softley
, and
B. R.
Heazlewood
,
J. Phys. Chem. A
119
(
50
),
12449
(
2015
).
45.
P. C.
Schmid
,
J.
Greenberg
,
M. I.
Miller
,
K.
Loeffler
, and
H. J.
Lewandowski
,
Rev. Sci. Instrum.
88
(
12
),
123107
(
2017
).
46.
D.
Rösch
,
H.
Gao
,
A.
Kilaj
, and
S.
Willitsch
,
EPJ Tech. Instrum.
3
(
1
),
5
(
2016
).
47.
S.
Jyothi
,
K. N.
Egodapitiya
,
B.
Bondurant
,
Z.
Jia
,
E.
Pretzsch
,
P.
Chiappina
,
G.
Shu
, and
K. R.
Brown
,
Rev. Sci. Instrum.
90
(
10
),
103201
(
2019
).
48.
M.
Hejduk
and
B. R.
Heazlewood
,
Rev. Sci. Instrum.
90
(
12
),
123701
(
2019
).
49.
P.
Micke
,
J.
Stark
,
S. A.
King
,
T.
Leopold
,
T.
Pfeifer
,
L.
Schmöger
,
M.
Schwarz
,
L. J.
Spieß
,
P. O.
Schmidt
, and
J. R.
Crespo López-Urrutia
,
Rev. Sci. Instrum.
90
(
6
),
065104
(
2019
).
50.
A.
Tsuchiya
,
X.
Lin
,
H.
Takayama
,
M.
Saito
, and
M.
Xu
, “
Development of 1.5 W 4 K two-stage pulse tube cryocoolers with a remote valve unit
,” in
Proceedings of the 18th International Cryocooler Conference
, Boulder, CO,
2014
, pp.
1549
1757
, ISBN 978-1-934021-04-0, ISSN 1549-1757, https://www.shicryogenics.com/wp-content/uploads/2021/02/Development-of-1.5W-4K-Tsuchiya-ICC-18.pdf.
51.
T.
Tomaru
,
T.
Suzuki
,
T.
Haruyama
,
T.
Shintomi
,
N.
Sato
,
A.
Yamamoto
,
Y.
Ikushima
,
R.
Li
,
T.
Akutsu
,
T.
Uchiyama
 et al,
Cryocoolers 13
(
Springer
,
2005
), pp.
695
702
.
52.
J. R.
Heim
,
Cryogenics
12
(
6
),
444
(
1972
).
54.
W. C.
Wiley
and
I. H.
McLaren
,
Rev. Sci. Instrum.
26
(
12
),
1150
(
1955
).