Fixed-target sample delivery device for protein x-ray crystallography

Abstract

A fixed-target device for serial crystallography of protein samples. The device comprises an inlet configured to receive a sample; and a main channel comprising cyclic olefin copolymer (COC), the main channel in fluid communication with the inlet, the main channel including a plurality of traps, each trap configured to trap a single crystal from the sample to obtain diffraction data of the single crystal upon application of X-ray to the main channel.

Description

FIXED-TARGET SAMPLE DELIVERY DEVICE FOR PROTEIN X-RAY CRYSTALLOGRAPHYCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application is a non-provisional of and claims the benefit of U.S. Provisional Application No. 63 / 664,668, filed on June 26, 2024, the contents of these applications being incorporated herein by reference.STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

[0002] This invention was made with government support under 2153503 awarded by the National Science Foundation. The government has certain rights in the invention.BACKGROUND

[0003] Macromolecular X-ray crystallography (MX) stands as a cornerstone technique in the field of structural biology, providing invaluable insights into the three-dimensional structures of proteins with atomic resolution. MX relies on the principle of X-ray diffraction originating from atoms within the crystal lattice generating a diffraction pattern. By analyzing this diffraction pattern, the spatial arrangement of atoms within the protein reveals its structural details. MX has revolutionized our understanding of protein structure and function, playing a central role in drug discovery, rational protein engineering, and the elucidation of biochemical mechanisms. By providing detailed insights into the atomic-level organization of proteins, MX continues to drive advances in biochemistry, molecular biology, and drug design, shaping our understanding of life at the molecular level.

[0004] The delivery of protein crystal samples into the path of an X-ray beam is a critical step that directly impacts the quality and reliability of the structural data obtained. Sample delivery approaches need to be adapted to the experimental conditions, such as cryogenic- temperatures, vacuum chambers or ambient pressure conditions where chambers are filled with helium gas to reduce background scattering. Protein crystals are mounted on goniometer pins with loops or caps and then positioned in the X-ray beam manually or with a robot. This ensures precise alignment for optimal data collection and supports cryogenic conditions. The advent ofX-ray free electron lasers (XFELs) enabled room -temperature serial crystallography (SX), requiring new sample delivery methods as crystals are destroyed after a single X-ray shot. In liquid delivery methods, protein crystals are delivered to the X-ray beam either through a continuous or segmented liquid jet. These methods include gas dynamic virtual nozzles (GDVNs), droplet-based injectors, microfluidic electrokinetic sample holders and high viscosity extruders. In fixed-target sample delivery systems, protein crystals are mounted in so-called fixed targets, such as silicon chips or polymer-based microfluidic devices, and then mounted on suitable holders for X-ray diffraction analysis. Silicon fixed-target chips were developed by etching materials to create windows that localize crystals, offering the advantage of minimal background scattering. However, precise positioning of these windows in the X-ray beam is crucial to prevent scattering from silicon and potential detector damage.

[0005] Accordingly, it would be desirable to have a fixed-target device for sample analysis and use with compact X-ray sources such as the compact X-ray light source (CXLS).SUMMARY

[0006] In an embodiment, the present disclosure provides a fixed-target device for serial crystallography. The device comprises an inlet configured to receive a sample, and a main channel comprising a thermoplastic material, the main channel in fluid communication with the inlet, the main channel including a plurality of traps, each trap configured to trap a single crystal from the sample to obtain diffraction data of the single crystal upon application of X-ray to the main channel.

[0007] In another embodiment, the present disclosure also provides a method of conducting serial crystallography. The method comprises delivering a sample to an inlet of a fixed-target device, applying vacuum to an outlet of the fixed-target device, trapping a plurality of crystals in the sample in a plurality of traps of the fixed-target device, applying X-ray from a compact X-ray light source to the plurality of traps, and generating diffraction data for each of the crystals.

[0008] Other aspects of the present disclosure will become apparent by consideration of the detailed description and accompanying drawings.BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

[0010] The features and advantages of the present disclosure, and the manner of attaining them, will become more apparent and the present disclosure will be better understood by reference to the description of the present disclosure taken in conjunction with the accompanying drawings, wherein:

[0011] FIG. 1 illustrates (a) a schematic of a fixed-target device according to an embodiment of the present disclosure, (b) enlarged view of a section of the device showing trap geometry, (c) SEM image of a KOH etched Si wafer with angled sidewalls at etched features, (d) an image of a sealed fixed-target device after filling with IPA dye solution shows accomplished device sealing, (e), (f), and (g) schematic representation of device fabrication where (e) imprinting of ~120 pm COC with an etched Si wafer, (f) solvent treatment of ~120 pm COC cover slide in a glass Petri dish, and (g) bonding of imprinted COC layer and COC cover slide.

[0012] FIG. 2 illustrates a schematic of the fixed-target device shown in FIG. 1 with an enlarged view of a section of the device showing trap geometry with X-ray beam incidence.

[0013] FIG. 3 is an image of cuboid lysozyme crystals of ~15 pm x 10 pm x 10 pm size grown using the batch method.

[0014] FIG. 4 illustrates a schematic image of the geometry used for the particle trapping simulation. Shown in a-c) are two geometric traps each, (a) The blue lines in between the posts of a trap were introduced to freeze particles representing a particle trap, (b) The green lines around the posts represent an outer boundary of the trapping structure to create a realistic particle migration pathway. These boundaries were set to the “bounce” condition, (c) The red lines in between two outer boundaries represent walls, set to the “bounce” condition which is only effective when the trap structure is pre-occupied by another particle.

[0015] FIG. 5 illustrates trapping efficiency of the fixed-target device with (a) 45 pm beads at 50 mbar, (b) 45 pm beads at 10 mbar and (c) two differently shaped lysozyme crystals at 50 mbar vacuum pressure applied to the outlet. Images of traps occupied with (d) cuboid lysozymecrystal (e) needle-shaped lysozyme crystals (f) 45 pm polystyrene beads; Scale bars are 400 pm; Before redistribution signifies right after devices were filled by applying 50 mbar pressure.After redistribution refers to the same device as initially filled but in addition after applying the redistribution sequence twice.

[0016] FIG. 6 graphically illustrates velocity of 45 pm polystyrene beads inside the fixed- target device at different vacuum pressures applied to the outlet of the devices.

[0017] FIG. 7 graphically illustrates Si wafer feature height as a result of KOH etching for 30 min, 45 min, and 50 min KOH etching time indicating an etch rate of 1.1 ± 0.09 pm / min. The linear regression resulted in a slope of 1.1 pm / min.

[0018] FIG. 8 graphically illustrates (a) 0cof the native COC surface and change in 0cfor 20 s plasma treated COC surface over 5 days, (b) chemically treated surfaces before (a) and after (b) washing with DI water.

[0019] FIG. 9 graphically illustrates percentage of liquid loss from two fixed-target devices obtained at three different time points after placing inside a vacuum chamber at 10'5mbar pressure. If not apparent, the error bars are smaller than the symbols.

[0020] FIG. 10 illustrates representative image of device section with needle-shaped lysozyme crystals (a) before placing in the vacuum chamber and (b) after placing in the vacuum chamber at 10'5mbar for 5 h.

[0021] FIG. 11 graphically illustrates numerical simulation result of trapping efficiency (a) 1st generation fixed-target design for 10 run cycles each cycle was simulated for 10 s. (b) 1st generation device with varying distances “g” for 60 s. (c) 2nd generation design with g = 95 and 25 pm, run cycle duration was 10 s.

[0022] FIG. 12 is a photograph of the fixed-target device (shown in red square) mounted atID29 beamline at the ESRF.

[0023] FIG. 13 graphically illustrates radial intensity average with corresponding radial intensity vector to quantify background contribution from a —240 pm COC device fdled with an aqueous solution.

[0024] FIG. 14 illustrates representative diffraction pattern from a 15 pm x 10 pm x 10 pm size lysozyme crystal. A) Squares on the image represent identified reflections (hits) by Peakfinder8 algorithm using SNR 3000, threshold 5, and 2 pixels per peak to identify a minimum of 10 peaks were used as hit finding parameters. B) Indexed pattern shown in A). C and D) A closer view of the black boxed area shown in A). Diffraction was seen beyond 2A resolution. E) Unit cell dimensions distribution from 7692 indexed patterns.

[0025] FIG. 15 illustrates lysozyme structure, (a) Overview of the 2mFo-DFc electron density maps (gray mesh) contoured at 1 G for the lysozyme structure (orange sticks), (b) A closer view of the residues at the catalytic site.

[0026] FIG. 16 illustrates a) 2mFo-DFc electron density maps around the disulfide bonds (Cys94-Cys96 (top), Cys66-Cys80(middle), and Cys6-Cysl27 (bottom)), b) Simulated annealing mFo-DFc omit maps around the residues and the disulfide bonds (Cys94-Cys96 (top), Cys66- Cys80(middle), and Cys6-Cysl27 (bottom)). Omit maps are contoured at 3G.

[0027] FIG. 17 illustrates superposition of the lysozyme structure presented in this study (orange) to other lysozyme structures determined using serial crystallography at ambient temperature (PDB 4RLM (green), 7S4W (magenta), and 4X3B (cyan).

[0028] Those of ordinary skill in the art will understand that the devices and methods specifically described herein and illustrated in the accompanying drawings are non-limiting exemplary embodiments and that the scope of the various embodiments of the present disclosure is defined solely by the claims. The features illustrated or described in connection with one exemplary embodiment may be combined with the features of other embodiments. Such modifications and variations are intended to be included within the scope of the present disclosure.DETAILED DESCRIPTION

[0029] Before any embodiments of the invention are explained in detail, it is to be understood that the invention is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the following drawings. The invention is capable of other embodiments and of being practiced or of being carried out in various ways.

[0030] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. In case of conflict, the present document, including definitions, will control. The materials, methods, and examples disclosed herein are illustrative only and not intended to be limiting. Methods and materials similar or equivalent to those described herein can be used in practice or testing of the disclosed invention. All publications, patent applications, patents and other references mentioned herein are incorporated by reference in their entirety.

[0031] The terms “comprise(s),” “include(s),” “having,” “has,” “can,” “contain(s),” and variants thereof, as used herein, are intended to be open-ended transitional phrases, terms, or words that do not preclude the possibility of additional acts or structures. The singular forms “a,” “and” and “the” include plural references unless the context clearly dictates otherwise. The present disclosure also contemplates other embodiments “comprising,” “consisting of’ and “consisting essentially of,” the embodiments or elements presented herein, whether explicitly set forth or not.

[0032] For the recitation of numeric ranges herein, each intervening number therebetween with the same degree of precision is explicitly contemplated. For example, for the range of 6-9, the numbers 7 and 8 are contemplated in addition to 6 and 9, and for the range 6.0-7.0, the number 6.0, 6.1, 6.2, 6.3, 6.4, 6.5, 6.6, 6.7, 6.8, 6.9, and 7.0 are contemplated, and for the range 1.5-2, the numbers 1.5, 1.6, 1.7, 1.8, 1.9, and 2 are contemplated.

[0033] The modifier “about” used in connection with a quantity is inclusive of the stated value and has the meaning dictated by the context (for example, it includes at least the degree of error associated with the measurement of the particular quantity). The modifier “about” shouldalso be considered as disclosing the range defined by the absolute values of the two endpoints. For example, the expression “from about 2 to about 4” also discloses the range “from 2 to 4.” The term “about” may refer to plus or minus 10% of the indicated number. For example, “about 10%” may indicate a range of 9% to 11%, and “about 1” may mean from 0.9-1.1. Other meanings of “about” may be apparent from the context, such as rounding off, so, for example “about 1” may also mean from 0.5 to 1.4.

[0034] The above-noted challenges, coupled with complex fabrication and issues like sample dehydration, has led to the development of thermoplastic and polymeric materials for fixed- target sample delivery including cyclic olefin copolymer (COC), cyclic olefin polymer (COP), Kapton, Mylar, polycarbonate (PC), poly(methyl methacrylate) (PMMA), and SU-8. These polymers offer low vapor permeability, minimal scattering, and easy fabrication, with some incorporating graphene layers to enhance performance.

[0035] In general, fixed-target systems enable rapid screening and data collection from multiple crystals without individual mounting, integrating easily with automated workflows to save time and boost productivity. They support in situ crystallization, minimize handling of fragile crystals, making it a valuable approach in structural studies, and achieve high hit rates with minimal sample consumption, scanning nearly 100% of crystals in the pulsed X-ray beam. Fixed-targets also allow multiple exposures of the same crystal, beneficial for low-fluence X-ray sources or weakly diffracting samples, enabling data summation for indexing and damage-free extrapolation in time-resolved studies.

[0036] The present disclosure provides a fixed-target device using thermoplastic material, such as, COC to meet the sample delivery needs of CXLS. The fixed-target device traps crystals in predefined positions, enabling multiple X-ray exposures and ensuring compatibility with vacuum conditions to prevent dehydration. Trapping efficiencies were tested with beads, crystals, and numerical modeling. Serial crystallography experiments at ESRF’s ID29 beamline using lysozyme microcrystals validated the device. A high-quality lysozyme structure confirmed its suitability for serial delivery at the CXLS and synchrotrons, advancing room-temperature protein crystallography.

[0037] The present disclosure provides a microfluidic device that can screen protein crystallization conditions, and is fabricated from two different polymers, cyclic olefin copolymer (COC) and poly dimethyl siloxane (PDMS). COC is used in the fluidic and actuation layer, as it has better x-ray transmission properties. PDMS is used for the elastomer membrane to allow valve integration. This device also screens crystallization conditions and can be used to directly carry out protein crystallography.

[0038] The present disclosure provides a fixed-target microfluidic system for delivering protein microcrystals to X-ray beams for diffraction data collection and structure determination. In some embodiments, the fixed-target system includes three symmetric sections arranged in an area of 1 in. * 1 in. with up to 18,000 crystal traps per device. In some embodiments, each trap is targeted to hold one crystal up to 50 pm in size in the largest dimension. The device has been fabricated using COC for high-quality diffraction data collection with low background scattering induced through the fixed-target material. In some embodiments, the fixed-target device is configured for vacuum compatibility which enables the use in vacuum experimental chambers of X-ray radiation sources including the CXLS. In some embodiments, image data (e.g., microscopy generated images) can be collected to identify the trapping location of the crystals in the device.

[0039] In exemplary embodiments, to assess the validity of the fixed-target device, serial crystallography experiments were performed on the model protein lysozyme at the new European Synchrotron Radiation Facility -Extremely Brilliant Source (ESRF-EBS) beamline ID29. A 1.6 A crystal structure of the protein was solved, demonstrating that, in general, the fixed-target device can be used to generate high-quality data from macromolecular crystals at the CXLS and synchrotron radiation sources, which holds enormous potential for advancing the field of protein structure determination by fixed-target X-ray crystallography.

[0040] FIGS. 1 (at a) and 2 illustrate a fixed-target device 100 according to an embodiment of the present disclosure. The fixed-target device 100 is formed on a substrate (discussed below) and includes a main channel 104 including a plurality of traps 108 configured to trap or secure an object, such as a crystal C in a sample. In some embodiments, the main channel 104 includes a depth of about 30 pm - 50 pm. In some embodiments, the depth of the main channel 104 can beup to 500 pm. The fixed-target device 100 includes an inlet 112, a plurality of inlet channels 116, an outlet 120, and a plurality of outlet channels 124. The inlet 112 is in fluid communication with the plurality of inlet channels 116, and the inlet channels 116 are in fluid communication with the main channel 104. The main channel 104 is also in fluid communication with the plurality of outlet channels 124, and the plurality of outlet channels 124 are in fluid communication with the outlet 120.

[0041] In some embodiments, the substrate can include a plurality of fixed-target devices 100. In the embodiment shown in FIG. 1 (at a), the substrate includes three fixed-target devices 100. In other embodiments, the substrate can include more than three fixed-target devices 100. In some aspects, each fixed-target device 100 includes an about 250 pm - 750 pm wide inlet 112 and outlet 120, branched into eight equally spaced 125 pm - 375 pm inlet channels 116 and outlet channels 124. In some aspects, each fixed-target device 100 includes a 250 pm wide inlet 112 and outlet 120, branched into eight equally spaced 125 pm inlet channels 116 and outlet channels 124. In some aspects, each fixed-target device 100 includes a 500 pm wide inlet 112 and outlet 120, branched into eight equally spaced 250 pm inlet channels 116 and outlet channels 124. In some aspects, each fixed-target device 100 includes a 750 pm wide inlet 112 and outlet 120, branched into eight equally-spaced 375 pm inlet channels 116 and outlet channels 124. These inlet channels 116 and outlet channels 124 are open to the main channel 104 which includes the plurality of traps 108 to hold crystals from a sample in specific positions. In some aspects, the main channel 104 is about 15 - 20 mm long and about 5-10 mm wide. In some aspects, the main channel 104 is about 15 mm long and about 5 mm wide. In some aspects, the main channel 104 is about 17 mm long and about 7 mm wide. In some aspects, the main channel 104 is about 20 mm long and about 10 mm wide. In some embodiments, the overall dimension of the fixed-target device 100 is about 2.0 - 3.0 cm x 2.0 - 3.0 cm. In some embodiments, the overall dimension of the fixed-target device 100 is about 2.0 cm x 2.0 cm. In exemplary embodiments, the overall dimension of the fixed-target device 100 is 2.5 cm x 2.5 cm. In some embodiments, the overall dimension of the fixed-target device 100 is about 3.0 cm x 3.0 cm. In some embodiments, each of the main channels 104 includes about 400 - 500 traps 108. In exemplary embodiments, each of the main channels 104 includes 468 traps 108 thereby creating 1,404 predefined positions for crystals to be probed by an X-ray beam as shown in FIG. 2.

[0042] With reference to FIGS. 1 (at b) and 2, the plurality of traps 108 are oriented in an array having a plurality of rows that are offset relative to adjacent rows. Each of the traps 108 include a first post 130, a second post 134, and a recess 138. In some examples, the recess 138 is defined by a first angled surface of the first post and a second angled surface of the second post. In some examples, the angled surfaces are not required depending on an ion etching process used. The recess 138 includes an opening 142 at a first end of the trap 108 that has a width or diameter configured to receive or trap a crystal. The opening 142 has a reduced width or diameter forming a bypass channel 146 that extends to a second end and opposite end of the trap 108. In some examples, the traps 108 have a geometry as shown below in Table 1 :TABLE 1EXAMPLE

[0043] As depicted in the Examples below, in certain exemplary embodiments, a fixed-target device is fabricated according to the following methods. It will be appreciated that, although the methods depict the fabrication of an exemplary fixed-target device, the following methods may be modified and / or result in the fabrication of a fixed-target device modified from the exemplary embodiment.

[0044] Fabrication of Fixed-Target Devices

[0045] A photomask design was created in AutoCAD 2023 (Autodesk) with the design shown in FIG. 1 (at a) and a film photomask was obtained from Great Lakes Engineering with 25,000 dpi resolution. Each mask contained the fixed-target design which was separated into three identical sections. Si wafers with 1000 nm SiCh coating were cleaned with acetone and isopropyl alcohol (IP A), followed by drying with a stream of N2. The Si wafers were then heated on a hot plate at 150 °C for 10 min. After cooling for 5 min, the wafers were spin-coatedusing 3 mL positive photoresist at 1000 rpm for 45 s. Spin-coated wafers were soft baked at 105 °C for 1 min and exposed at an average of 150 mJ / cm2at a wavelength of 365 nm using a Suss MJB4 mask aligner (Karl SUSS, Germany). Following the exposure, the wafers were baked at 115 °C for 1 min before developing them in 0.2 M tetra methylammonium hydroxide (TMAH) for 1 min.

[0046] The Si wafers were then immersed into buffered hydrofluoric acid (HF) at room temperature for 10 min to remove the 1000 nm SiCh layer. The HF-etched wafers were thoroughly washed with deionized (DI) water which was followed by another washing step with a 1 : 1 mixture of IPA and acetone. The wafers were dried in a stream of N2 before storing them at room temperature. An aqueous solution of 50 % potassium hydroxide (KOH) and 10 % IPA was prepared to further etch the HF-etched Si wafer. A KOH solution temperature ranging from 76 to 78 °C was maintained throughout the etching cycle and the wafers were etched for various durations to determine the etch rate. The KOH-etched wafers were subsequently washed with DI water and feature heights were measured using an Olympus BX53M microscope (Olympus, Japan).

[0047] Following fabrication, the wafers were silanized by placing them in a desiccator along with 100 pL of (Tridecafluoro- 1,1, 2, 2-tetrahydrooctyl) trichlorosilane (TDTS) for 1 h before they were used for imprinting. FIG. 1 (at e-g) shows a schematic representation of the device fabrication steps. First, the 175 pm thick COC layers were sandwiched in between two 125 pm Kapton foil sheets (American Durafdm, USA). The entire assembly was placed in between two platens of a Carver benchtop hot press (model 5420, Carver Inc., USA) at 202 °C, and a pressure of 600 psi was applied for 15 min after the set temperature was achieved. For imprinting, the resulting COC film of about 120 pm thickness was placed on the etched Si wafer and sandwiched in between a 10 cm glass wafer and a 125 pm Kapton foil. Another 10 cm diameter glass wafer was placed at the back of the Kapton foil, and the entire assembly was placed in the hot press. A temperature of 202 °C and 150 psi pressure were then applied for 10 min. The hot press was cooled down for 10 min and then pressure was released. The resulting imprinted COC was removed from the Si wafer and a microdrill (Cameron Microdrill Press, USA) was used to drill holes of 1.5 mm diameter at the positions of the inlets and outlets. After drilling holes, the COC pieces were cut into 3 cm x 3 cm slabs and sonicated in IPA for 15 minto clean them. Another featureless piece of COC with about 120 pm thickness was used to seal the imprinted layer of COC. The featureless layer was placed on the air interface of a 65 % ethanol and 35 % cyclohexane(vZv) mixture for 5 min and then placed on a glass slide with the treated side facing up. The cover slide layer was dried with N2 and the imprinted COC layer was placed on the solvent-treated COC slab. The two layers were manually pressed together to ensure contact and the assembled devices were stored in air (> 12h) until further use.

[0048] Etched wafers were sputter-coated with gold and imaged with a Zeiss Auriga Crossbeam (Carl Zeiss AG, Germany) scanning electron microscope equipped with an Everhart- Thornley detector. The images were collected in 84x magnification at a tilt angle of 45° using a beam voltage of 20 kV.

[0049] Lysozyme crystals were prepared in two size ranges. A 40 mg / mL lysozyme solution in 0.5 M sodium acetate buffer (pH 3.5) and the precipitant solution 20% w / v NaCl and 6% w / v PEG 6000 were prepared. Small cuboid crystals (about 15 pm x 10 pm x 10 pm) were grown overnight at room temperature using the batch method by mixing 1 mL of the precipitant with 175 pL of the protein solution (see FIG. 3). For larger crystals, both solutions were diluted by half, producing needle-like crystals (—60 pm x 6 pm x 6 pm). Only the cuboid microcrystals were used in the serial diffraction experiments.

[0050] Polystyrene beads of 45 pm diameter were used to demonstrate the captureability of crystals inside the traps of the fixed-target devices. Devices were filled by pipetting 10 pL of IPA to the inlet reservoirs filling the channels by capillary action. A negative pressure of 50 mbar was then applied to the outlet reservoir through a pump (MFCS-EZ, Fluigent, France) to replace the IPA with an aqueous solution containing Fl 08 to avoid bead clogging. The pump was connected to the outlet reservoir by Tygon tubing. Once the liquid meniscus decreased below the level of the inlet reservoir, 5 pL of bead suspension was added to the inlet. During the filling procedure, the fluid in the inlet reservoir was continuously agitated with a pipette tip to prevent inlet clogging. Pressure was applied to the outlet until bead trapping was observed in the last row of the device section which took about 60 s. Following trapping, the channel was washed with 5 pL water to remove any beads that were not trapped. The entire filling process was observed under a stereomicroscope (Olympus SZ61, Japan) while the stage was illuminatedby a Schott ACE1 gooseneck lamp. Moreover, each trap of the devices was inspected under the microscope to find the bead occupancy. The trapping efficiency (TE) of the device was calculated using the method described in the Supporting Information. To redistribute the polystyrene beads throughout the channels, 50 mbar negative pressure was applied to the inlet for 2 s and subsequently to the outlet for 10 s. This procedure was repeated 10 times before determining TE again. The redistribution sequence was also performed with polystyrene beads at 10 mbar pressure.

[0051] Similarly, for lysozyme crystals, the channels were first filled with IPA and then 5 pL of crystallization buffer was added to the inlets once the meniscus of IPA decreased below the top surface of the inlet reservoirs. Then, 4 pL of small lysozyme crystal slurry were added to the inlet and 50 mbar negative pressure was applied to the outlet the same way as described above for the beads. Once the liquid meniscus decreased below the level of the inlet reservoir, 4 pL crystallization buffer was pipetted into the inlet reservoir to prevent it from drying out. Also, the inlet reservoir was continuously agitated with a pipette tip to prevent the crystals from accumulating and clogging the inlet. The Tygon tubing was removed from the outlet reservoir to stop applying the negative pressure once crystals occupied the last row of traps of the device. For needle-shaped lysozyme crystals, 5 pL crystal slurry was added after fdling the device with buffer as above. The same pressure was applied to the outlet while adding another 5 pL of crystal suspension. The channels were washed with a total of 30 pL buffer solution after the second round of crystal slurry addition. The traps were checked under the stereomicroscope to find the number of traps occupied with crystals following the same procedure as outlined above. To evenly distribute the crystals throughout the channels, negative pressure was applied to the inlet for 2 s and subsequently to the outlet for 10 s. This cycle was repeated 10 times and trap occupancy was determined by microscopy inspection to determine the TE.

[0052] Trapping Efficiency (TE)

[0053] COMSOL Multiphysics 6.2 (COMSOL Inc, MA, USA) finite element analysis software was used to find the theoretical trapping efficiency, TE, of the crystals in the fixed- target devices. The geometry of the traps (as shown in FIGS. 1 (at b) and 2) representing the device was imported into COMSOL from AutoCAD. Then, the Creeping Flow module wassolved to obtain the convective velocity field for the pressure-driven flow. In this module, the Navier-Stokes equation was solved in a simplified form as described below: / zV2u — Vp = 0 Eq. 1 where p is the dynamic viscosity of the medium, u is the fluid velocity and p is the pressure. The inlet boundary condition was set to the flow velocity which was calculated experimentally for pressure values of 10 mbar, 20 mbar, 30 mbar, and 50 mbar to match the simulation condition to that of the experimental condition (see FIG. 6 for the resultant values and “Bead velocity determination”). The channel walls and the post walls were assigned a no-slip (u = 0) boundary condition, while the outlet was designated as an open boundary. The fluid properties of water were used as system parameters. Next, the Particle Tracing Module was solved to obtain the force acting on the particles and their trajectories solving:F = 6npru Eq. 2 where F is the force acting on the particles, r is the radius of a particle (= 45pm) and u was obtained from the first study. This model was used to determine the flow trajectories of particles after being released from the inlet and to determine the number of particles trapped.

[0054] Additional wall boundaries were added to the physical geometry of the device to account for the particle trapping occurring experimentally. A line (represented in FIG. 4 (at a) in blue) was established spanning the distance forming the bypass set to properties of a wall that freezes particles and holds them after they reached this position. To simulate realistic particle migration along the microfluidic channel geometry, an outer wall boundary around the trapping structure (represented in FIG. 4 (at b) in green) was added taking into account the particle diameter. This boundary was placed around each trap at a distance equaling the particle radius. Additionally, a third line was added as shown in FIG. 4 (at c) (red lines) to close a trap thus simulating the hydraulic resistance for an occupied trap. The trapping efficiency (ZE) of each device section was found by using the ratio of the number of traps occupied (No) over the total number of traps (NT), expressed as a percentage:

[0055] One of the three sections was assessed for each case to determine the TE. Occupied traps were identified through trapped beads visually and counted manually.

[0056] Devices were inspected with an upright microscope (Olympus BX53M microscope, Olympus, Japan) after filling experiments, and occupied traps were counted. Representative images of occupied traps are shown in FIG. 5 (at d-f). Measurements were performed in one section out of three from three different devices.

[0057] Bead Velocity

[0058] The migration of 45 pm beads during the filling process at vacuum pressures of 10, 20, 30, and 50 mbar was recorded by video microscopy in the first-generation fixed-target device and the velocities were determined by particle tracking analysis. In brief, the particles recorded were tracked using ImageJ manually and the distance the particle traveled was calculated. Then, the traveled distance was divided by the time required for this migration distance to calculate the velocity at different pressures (summarized in FIG. 6). The obtained experimental particle velocity was used in the numerical model.

[0059] Vacuum Compatibility

[0060] After filling the complete devices with crystals, the inlets and outlets were sealed with 10 min epoxy glue. The devices were stored for 24 h at room temperature before performing vacuum compatibility tests. Weight measurements were performed before placing the devices inside a vacuum chamber. To assess the resilience of the fixed target device within a high vacuum environment (<IO'~ mbar), the devices underwent multiple placements in such conditions - a notable challenge due to the time-consuming process of spinning down and venting the chamber associated with the requisite turbomolecular pump. To address this challenge and facilitate rapid testing and replacement of the device the primary chamber was interlinked with an ultrahigh vacuum (UHV) chamber situated beneath it, with the two sections separated by a gate valve. The main chamber utilized a roughing pump (XDS35i dry scroll pump, Edwards Ltd., U.K.) while the UHV chamber employed a turbomolecular pump (Turbo V 301 NAV, Agilent) supported by an additional roughing pump (XDS35i dry scroll pump, Edwards Ltd., U.K.). Prior to testing, the UHV chamber underwent pre-evacuation to highvacuum, with the gate valve closed. During the loading phase, the devices were introduced into the main chamber at atmospheric pressure through a vacuum door. Subsequently, the main chamber was evacuated to a rough vacuum (about 10'1mbar) using its designated roughing pump. As the pressure reached an optimal level, the gate valve was opened, while the main vacuum chamber remained isolated from the roughing vacuum pump. For the removal or replacement of the device in a vacuum, the gate valve was securely closed, and the upper chamber was vented to facilitate the opening of the main door, providing access to the device within the main chamber. The vacuum chamber pressure was monitored using 972B Pressure Transducers (MKS).

[0061] The mass of the devices was measured at 1, 1.5, and 2.5 h after placing them in the vacuum chamber with a Mettler Toledo UMX2 microbalance (Mettler Toledo, Switzerland). The amount of liquid loss over time was calculated by subtracting the mass at each time point (Wt) from the initial device mass (Wi). This was expressed as % liquid loss at each assessed time point (%L): 100 Eq. 4

[0062] The total liquid mass (Lt) in the device was determined to amount 19.17 mg by calculating the volume of the channel sections and assuming a density of water of 0.9982 g / mL.

[0063] X-Ray Scattering and Diffraction

[0064] Diffraction experiments were performed at the ID29 beamline at the ESRF-EBS (Grenoble, France) using a 1% bandwidth X-ray beam focused to 4 pm * 2 pm (H * V) (fwhm) with a pulse length of 90 ps at a repetition rate of 231.25 Hz. (66,67) A -240 pm thick COC fixed-target microfluidic chip was mounted on a holder frame of 3 cm x 3 cm and scanned using an S-shaped trajectory with an X-ray energy of 11.56 keV (wavelength of 1.072 A). Diffraction frames were recorded every 30 pm with a vertical spacing of 30 pm using a 90 ps X-ray pulse. Data acquisition was performed with a MD3upSSX diffractometer and X-ray diffraction data were recorded using a Jungfrau 4M detector placed at a distance of 99 mm from the fixed-target chip. Data collection statistics are listed in Table 2.TABLE 2

[0065] To quantify the X-ray scattering background from the fixed-target device, radial intensity (7) as a function of the radial integrated scattering vector (q) was plotted using MATLAB 2023a (Mathworks). The MATLAB script was developed to group pixels at certain distances from the beam center and then to calculate the average intensity of those pixels in each group. The distance value for each group was converted to the q value and was plotted against I. The equation used to find q iswhere 20 is the angle of diffraction and is the wavelength of the X-ray beam.

[0066] Data Processing and Structure Determination

[0067] All data processing was carried out remotely using the Virtual Infrastructure for Scientific Analysis (VISA, https: / / visa.esrf.fr) platform. Hit-finding was performed using a GPU version of NanoPeakCell implementing the Peakfmder8 algorithm. An SNR of 3000, a threshold of 5, and a minimum of 2 pixels per peak to identify a minimum of 10 peaks were used as hitfinding parameters. The Bragg reflections were integrated using the software package CrystFEL (version 0.10.1) after indexing was attempted with CrystFEL’s indexamajig using the algorithms MOSFLM, DIRAX, XDS, XGANDALF and ASDF, in that order. The intensities were integrated by applying radii of 4, 5, and 7 and merged into the point group lmmm, using the CrystFEL program partialator.

[0068] MTZ files for phasing and refinement were generated by the CTRUNCATE program from the CCP4 software package and a fraction of 5% reflections were included in the generated Rfree set. Initial phases were obtained by molecular replacement with MOLREP. The previously published serial crystallography structure (PDB 5UVI) was used for lysozyme as the search model. The obtained model was refined using alternate cycles of automated refinement with REFMAC5 and manual inspection was performed with COOT. The final refined structure was validated using the Protein Data Bank (PDB) validation service prior to deposition. The atomic coordinates and structure factors have been deposited in the PDB with accession code PDB 9FA5. The final refinement statistics of the protein are given in Table 2. 2mFo-DFc electron density maps and simulated annealing mFo-DFc omit maps were calculated with the MAPS toolin the PHENIX software suite. All structure figures presented in this manuscript were generated with PYMOL (version 2.4.1) (Schrodinger LLC).

[0069] RESULTS

[0070] The fixed-target devices 100 were developed for sample delivery to be compatible with a wide range of X-ray sources including CXLS. The fabrication procedures, the efficiency of the devices for crystal capture, the device vacuum compatibility, and the background scattering contribution from the COC material were optimized.

[0071] Device Fabrication: the devices 100 were configured to include three identical sections of a wide microfluidic channel in which geometric traps were designed to capture a crystal of a certain size (FIG. 1). The fixed-target device depth and the trap geometry influenced the crystal sizes that can be trapped. Etching procedures in Si were used from which the polymer could be hot-embossed to generate traps on the final COC fixed-target device.

[0072] An important design feature was ensuring that the liquid flows through the traps during device filling and not around them, to prevent any unoccupied traps. Thus, trap bypasses with a width, c, as defined in FIG. 1 (at b) were incorporated in the design with the purpose of reducing the hydraulic resistance. Since the Si patterns were fabricated with wet chemical etching, the top dimension of the trap bypass channel in the final fixed-target device was larger compared to the patterned mask layer as KOH etching occurs at an angle of 54.7° with respect to the

[0100] Si surface. An SEM image of a section of a wet etched Si wafer is shown in FIG. 1 (at c) where the trap bypass and the angled sidewalls of the trap features are depicted. The etching depth on the Si-wafer was assessed upon KOH treatment resulting from an etch rate of 1.1 ± 0.1 pm / min (see also FIG. 7) which is slightly lower than reported in the literature for the same KOH etching solution. This may be explained by a higher etch bath temperature of 80 °C employed by Williams et al. compared to a range of 76-78 °C used here. Thus, the Si master wafers used to fabricate the fixed-target devices had feature heights of —50 pm, leading to the same channel depth for the manufactured devices.

[0073] It is noteworthy that surface modification of COC was performed using oxygen plasma and chemical treatment as described below.

[0074] Water contact angle (0C) measurements of native COC and modified COC surfaces were performed using a homemade drop imaging setup. The setup included a digital microscope (Jiusion USB Digital Microscope, China), an iPhone 14 Pro (Apple Inc., USA) flash for background illumination, and a plastic box as the stage. COC slabs were treated by exposing them to oxygen plasma for 20 s in a plasma cleaner (Harrick, USA). Slabs of plasma-treated COC surfaces were further immersed in 1 mM aqueous solution of Fl 08, 1 M KOH, or 1 M HC1 overnight. To measure the water contact angles of these surfaces, 3 pL water droplets were placed on the surface and images were recorded. The corresponding contact angles were determined with the “Contact Angle” plugin in ImageJ software (NIH, USA). Subsequently, surfaces were washed with DI water, and measurements were repeated following the same procedure as above.

[0075] The contact angle, 0c, of the COC / water / air three-phase system was measured and resulted in a slightly hydrophobic surface with 0C= 92 ± 1.3°, which prevents fdling by capillary action. Surface modifications of COC were therefore performed to obtain hydrophilic surfaces facilitating filling with crystal suspension by capillary action. To render the surface hydrophilic, plasma oxidation was performed resulting in a change to a hydrophilic surface with 0Cof 37 ± 2.3° upon 20 s plasma oxidation. The achieved hydrophilicity after plasma treatment was not permanent as demonstrated by an increase in 0Cover time, as shown in FIG. 8 (at a). In order to obtain permanent hydrophilicity, plasma-treated COC surfaces were further chemically treated as described above. 0c values of 13 ± 4.4°, 38 ± 2.4° and 55 ± 0.4°, for treatments with 1 mM F108 solution, 1 M KOH solution, and 1 M HC1 solution, respectively, were obtained implying that the surfaces became hydrophilic. 0Cvalues increase to 58 ± 0.9°, 57 ± 1.5° and 79 ± 3° for the same surfaces after they are washed with DI water. These measurements are summarized in FIG. 8 (at b).

[0076] Although both the plasma and chemically treated COC surfaces become hydrophilic, the fully assembled COC devices could not be fdled by capillary action with water or buffer solutions. A potential factor responsible for this effect might be the bonding process where cyclohexane is used to soften the COC surface, even though this process was only performed on the flat COC cover. While the featureless cover slide might lose the surface hydrophilicity during solvent exposure due to polymer chain rearrangement, a similar change might occur to theimprinted side after the bonding process due to residual cyclohexane on the flat COC surface. Since the chemical treatments did not result in the expected improvements in the filling procedures by capillary action the following protocol was developed for device filling: first, the channels were filled with IPA by capillary action, which was then replaced with a buffer solution as described above. The devices were inspected for any flow of dye solution beyond the channel boundaries with a microscope. No apparent leakage was observed (see FIG. 1 (at d)) confirming that the developed fabrication method resulted in sealed fixed-target devices at room temperature and atmospheric pressure.

[0077] Crystal Trapping Efficiency: one motivation behind the device design was to hold crystals in predefined positions and to expose those locations to the X-ray beam to obtain at least one diffraction pattern per trapped crystal. To assess the device’s capability of trapping crystals, the TE was assessed as discussed above. The TE was studied by filling the devices with 45 pm polystyrene beads, cuboid lysozyme crystals, and needle-shaped lysozyme crystals. It is important to note that TE for cuboid lysozyme crystals was assessed with a channel depth of 30 pm whereas for the other two cases, a channel depth of 50 pm was used.

[0078] Trapping of 45 pm polystyrene beads at 50 mbar negative pressure applied at the outlet reservoir to induce flow yielded an overall TE of 97.8 ± 0.48% from three sections. To further improve the occupancy, redistribution attempts of captured beads were performed as described above. This resulted in a decrease of TE to 91.7 ± 9.7%, which was undesirable. However, the single occupancy of the traps with beads increased slightly from 1.1 ± 0.0 to 2.5 ± 1.1% (FIG. 5 (at a)) which is not a significant improvement.

[0079] As shown in FIG. 5 (at b), filling the fixed-target devices at five times lower pressure decreases the overall TE but increases the number of traps occupied with single beads.Redistribution of beads in these devices resulted in a minimal increase of traps occupied with single beads but decreased the overall TE significantly. Since 50 mbar negative pressure resulted in higher TE of beads, this pressure was used to assess the TE of the two lysozyme crystal samples differing in their shapes.

[0080] The TE of cuboid lysozyme crystals resulted in 71.5 ± 2.8% (see FIG. 5 (at c)). The TE of the devices decreased to 66.5 ± 2.8% after redistribution demonstrating a similar behavioras for the polystyrene beads. In the case of needle-shaped lysozyme crystals, the same experiment resulted in 66.8 ± 3.5% TE. In contrast to the previous two samples, the TE was further increased to 82.3 ± 1.6% by redistributing the crystals inside the device. This can be explained by the tendency of the needle crystals to form clusters inside the channels during the initial filling process whereas, during the redistribution process, the clusters separate into individual crystals increasing the number of occupied traps, and thus resulting in higher TE. Additionally, FIG. 5 (at d-f) displays occupied traps in a subsection of the device for the three samples discussed above.

[0081] Vacuum Compatibility: Typically, during a serial X-ray crystallography experiment, vacuum chambers are often used to mitigate background scattering due to air. The newly developed CXLS has been designed with a vacuum experimental chamber operating at a pressure of ~10'5mbar. Thus, the vacuum compatibility of the devices was assessed at similar chamber pressure as described above. Two fully assembled devices, ~240 pm thick, filled with an aqueous solution were placed inside the chamber for three different time points and the percentage of liquid loss (%L) was determined. FIG. 9 shows that there was no measurable loss of enclosed liquid up to 2.5 h whereas after 5 h liquid loss was apparent, however at <0.1%. In addition, devices filled with lysozyme crystals were subjected to a vacuum pressure of ~10'5mbar. Comparison between the crystal quality prior to placing the devices in a vacuum and after 5 h vacuum treatment indicates no discernible differences (FIG. 10). These results demonstrate that the devices can successfully prevent dehydration of the enclosed crystals in vacuum experimental chambers of X-ray instruments for up to 5 h during serial crystallography experiments.

[0082] Design Optimization Aided by Numerical Modeling: while the above studies revealed satisfactory trapping occupancy with polystyrene beads used to model protein crystals of similar size-ideally suited for the design dimensions of the obtained devices-the single bead trapping efficiencies resulted in values <40%, which is undesirable for serial crystallography experiments. Furthermore, a single first-generation (FIG. 1 (at a) device with three sections can hold a maximum of 1404 traps (468 traps per section) at full occupancy, which is borderline to obtain a complete data set. Improvements in the device design were thus further assessed withnumerical simulations modeling the filling and trapping process with polystyrene beads, as described above. The boundary conditions used for the numerical modeling are shown in FIG. 4.

[0083] First, the numerical model was validated by assessing the TE employing the flow rates and bead concentration used experimentally. FIG. 11 (at a) shows the numerical simulation result of TE for a fixed-target device where each cycle refers to a run time of 10 s. Comparison between the experimental device filling time (60 s) and the same numerical simulation time shows that for both cases the TE was >90%. This validates that the numerical simulation model can be used to further study the fixed-target device filling. It was noted that the experimentally resultant trap geometries varied from the ideal design simulated matching the CAD design of the photomask. This is due to the wet chemical etching conditions. In future iterations of these devices, deep reactive ion etching will be employed, in which nearly straight channel walls are obtained and the numerical model should represent the experimental conditions more closely.

[0084] Next, the horizontal distance between the center of a trap and the nearest one in the next row at distance g was varied from 0 to 95 pm (see FIG. 1 (at b)) for the definition of g) to assess the optimum trap placement resulting in the highest TE. The TEs of the designs varying in g were also obtained at different velocities mimicking different filling pressures. Intuitively, one would expect TE to increase with lower velocity since more time is available to interact with the traps and thus fill them. However, a decrease in TE was observed at a lower velocity (FIG. 11 (at b)), which supports the experimental observation of FIG. 5 (at a,b), where a low pressure was employed for filling. Furthermore, the simulation results show that the highest TE is obtained for a design with g = 25 pm for the highest velocity (FIG. 11 (at b)). Other g values between 15 and 65 pm also resulted in TE values above 55% and TEs were generally higher at faster velocities. This result implies that at g values ranging from 15-65 pm, the trap positions fall along the particle trajectories and thus yield higher TE values. However, the highest TE value was achieved for g = 25 pm and 0.014 m / s, which was subsequently used for further optimization studies.

[0085] Lastly, the trap density was improved by reducing the distances a, b, d, and f (FIG. 1 (at b)). This resulted in 6004 traps per section which corresponded to approximately a 13-fold increase over the first-generation design. The numerical simulation study of TE with increasedtrap density was then performed with g = 95 pm to correlate to the first-generation design and with 25 pm as this yielded in highest TE from FIG. 11 (at b). In both cases, FIG. 11 (at c) indicates that TE > 99% can be achieved. The design with g = 25 pm requires only three cycles to achieve this TE whereas the design with g = 95 pm requires 18 cycles to fill the same number of traps. The second-generation design with g = 25 pm should thus allow to fabricate trap-based fixed-target devices which will require less filling time to occupy most traps, achieve almost perfect TE, and require only — 10 pL sample volume.

[0086] X-ray Scattering Background: the X-ray scattering background of the newly developed COC devices was assessed at the ID29 beamline at ESRF-EBS during experiment MX-2539 (see FIG. 12 for a representation of the setup). As shown in FIG. 13, the COC X-ray scattering background appears as a high-intensity peak at a q value of 1.2 A'1which corresponds to -5.2 A in real space. This result matches the COC scattering background observations previously reported in the literature. The appearance of a low-intensity water peak at 3 A'1, which corresponds to a resolution of —2.1 A, originates from the buffer solution with which the devices were filled during the experiment matching with the reported values from the literature. It is noteworthy to mention that the thickness of the probed sample layer was 50 pm compared to the entire COC device thickness of 240 pm. Thus, a —190 pm thick COC layer around each well contributed to the X-ray background scattering. The background scattering profile and intensity originating from COC will ultimately determine if diffraction patterns from protein crystals in these devices can be obtained, which was further tested with serial crystallography of lysozyme as discussed below.

[0087] Crystal Structure of Lysozyme: serial data collection on lysozyme microcrystals was conducted at the ID29 beamline at ESRF-EBS using the COC devices presented herein. Diffraction images were collected across the three sections of the fixed-target devices as outlined above, with a sample-to-detector distance of 100 mm, corresponding to 1.4 A resolution at the detector edge (FIG. 14). A total of 10 pL of crystal slurry (two COC chips) was used for a complete data set, with 104,000 images recorded per chip. Lysozyme microcrystals were observed to diffract up to 1.4 A resolution and belonged to the space group P43212 with unit cell dimensions of a = 78.3 A, b = 78.3 A, c = 37.7 A and a = 0 = y = 90°. A representative diffraction pattern from a lysozyme microcrystal and the unit cell dimensions distribution areshown in FIG. 14. For structure determination, a total of 208,000 frames were collected from two COC chips, of which 70,193 were classified as hits by CrystFEL and 15% of the identified hits were indexed, giving rise to a total of 10,543 indexed, integrated, and merged frames. The structure was solved by molecular replacement using the PDB entry 5UVJ as a search model without water molecules and was refined at a final resolution of 1.6 A with Rwork and Riiee of 19.1 and 25.1%, respectively. All data collection and refinement statistics are listed in Table 2.

[0088] The quality of the lysozyme structure can be assessed from the electron density map 2mFo-DFc of the active site residues of lysozyme (FIGS. 15 (at a-b) and 16 (at a)). Since disulfide bonds are especially susceptible to radiation damage, simulated annealing mFo-DFc omit maps were generated for all four disulfide bonds and the surrounding residues (FIG. 16 (at b)). The electron density surrounding the disulfide bonds confirms the absence of detectable radiation damage in our structure. The quality of the structure was further evaluated by comparing it with previously reported crystal structures determined using serial crystallography at ambient temperature including the PDB entries 4RLM, 7S4W, and 4X3B. As can be seen in FIG. 17, overall, the lysozyme structures aligned very well with each other, with an average RMSD value of 0.260 A. The average RMSD values for all atoms (0.374) indicate slightly higher differences, which are mainly found in the loop regions, as well as in the solvent-exposed regions, as expected.

[0089] Accordingly, the present disclosure provides a highly efficient, low sample consumption, and easy-to-use sample delivery COC device that can maximize the potential of serial crystallography. The COC-based fixed-target device can trap crystals in ~1400 predefined positions per device, which can be exposed to X-ray beams for macromolecular structure determination. This functionality of the devices can enable precisely locating the crystals in the X-ray beam path toward reaching maximum hit rates as well as allowing multiple exposures on the same crystal. As part of the development, the fabrication procedure has been optimized to make reproducible devices for X-ray crystallography experiments in vacuum chambers of up to ICT5mbar. The devices have also been demonstrated to trap spherical beads of ~45 pm diameter with >90%, cuboid lysozyme crystals with >70%, and needle-shaped lysozyme crystals with >80% efficiency. Moreover, numerical modeling has improved the design increasing the numberof predefined positions to ~ 18,000 per device. Additional improvements to the design included positioning the traps to increase the TE of the devices by close to 100%.

[0090] The functionality of the COC devices toward SX was demonstrated by collecting diffraction data and structure determination from lysozyme at 1.6 A at the ESRF-EBS ID29 beamline. The devices presented herein are easy to handle for the experimenter once assembled. Filling procedures can be accomplished with readily available vacuum sources, and the overall design footprint can be easily adapted to other end stations since the devices are thin and can be easily cropped with scissors if required. Overall, this COC-based fixed-target device holds the potential to be used for SX at a wide range of X-ray sources, including the CXLS. This novel X- ray source hosts a vacuum experimental chamber, which makes the fixed-target device an ideal candidate for the commissioning experiments of CXLS.

[0091] Additional features and advantages of the present disclosure are provided in the following claims.

Claims

CLAIMS:What is claimed is:

1. A fixed-target device for serial crystallography, the device comprising: an inlet configured to receive a sample; and a main channel comprising a thermoplastic material, the main channel in fluid communication with the inlet, the main channel including a plurality of traps, each trap configured to trap a single crystal from the sample to obtain diffraction data of the single crystal upon application of X-ray to the main channel.

2. The fixed-target device of claim 1, further comprising a plurality of inlet channels in fluid communication with the inlet and the main channel.

3. The fixed-target device of claim 1 or 2, wherein the main channel includes a depth of about 30 pm to about 50 pm.

4. The fixed-target device of any one of claims 1-3, wherein the plurality of traps are oriented in an array having a plurality of rows.

5. The fixed-target device of claim 4, wherein each row in the plurality of rows is offset relative to an adjacent row in the plurality of rows.

6. The fixed-target device of claim 5, wherein the offset is between about 15-65 pm.

7. The fixed-target device of claim 6, wherein the offset is about 25 pm.

8. The fixed-target device of any one of claims 1-7, wherein the plurality of traps have a greater trapping efficiency when the sample moves through the main channel at a greater velocity.

9. The fixed-target device of claim 8, wherein trapping efficiency is greater than 55% when the offset is about 25 pm and the velocity is about 0.014 m / s.

10. The fixed-target device of any one of claims 1-9, wherein each trap of the plurality of traps includes a first post, a second post oriented relative to the first post to form a recess configured to trap a single crystal, and a bypass channel between the first post and the second post, the bypass channel extending a length of the first post and the second post.

11. The fixed-target device of claim 10, wherein the first post includes a first angled portion and the second post includes a second angled portion, and wherein the first post and the second post are oriented to form the recess.

12. The fixed-target device of claim 10 or 11, wherein the bypass channel has a width of about 15-30 pm.

13. The fixed-target device of any one of claims 10-12, wherein the recess defines an opening having a width of about 60-95 pm.

14. The fixed-target device of any one of claims 1-13, wherein the X-ray is generated by a compact X-ray light source.

15. The fixed-target device of any one of claims 1-14, wherein the plurality of traps includes at least 1,400 traps.

16. The fixed-target device of any one of claims 10-13, wherein the recess is configured to trap a single crystal having a sized of 50 pm in its largest dimension.

17. The fixed-target device of any one of claims 1-16, further comprising an outlet in fluid communication with the main channel, the outlet configured to be coupled to a vacuum source to control a velocity of the sample through the main channel.

18. The fixed-target device of any one of claims 1-17, wherein the thermoplastic material is selected from a group consisting of cyclic olefin copolymer (COC), cyclic olefin polymer (COP), Kapton, Mylar, polycarbonate (PC), poly(methyl methacrylate) (PMMA), and SU-8.

19. The fixed-target device of claim 18, wherein the thermoplastic material is cyclic olefin copolymer (COC).

20. A method of conducting serial crystallography, the method comprising: delivering a sample to an inlet of a fixed-target device of claim 1; applying vacuum to an outlet of the fixed-target device of claim 1; trapping a plurality of crystals in the sample in a plurality of traps of the fixed-target device of claim 1; applying X-ray from a compact X-ray light source to the plurality of traps; and generating diffraction data for each of the crystals.

Images