700 MPa sample stick for studying liquid samples or solid-gas reactions down to 1.8 K and up to 550 K

Abstract

A new sample stick for top-loading cryostats allowing to investigate liquid samples and solid-gas reactions with neutron scattering is described. This stick prevents the freezing of the fluids injected into the cryostat down to the sample cell. The high-pressure capillary of the stick is thermally isolated from the cryogens’ baths and heated to a desired temperature. The maximum pressure that can be applied remotely to the sample is 700 MPa and the sample temperature may be controlled between 1.8 and 550 K. This new stick allows for example to explore $(p, T)$ phase diagrams of biological samples that were inaccessible. Extended versions provide 2- or 4-wire in-situ complementary measurements.

Keywords

Neutron scattering sample environment high pressure low temperature biology solid-gas reaction

1. Introduction

To date many studies on thermally induced protein folding/unfolding exist but much less is known about high pressure or combined effects, essentially due to technological difficulties to build up such experiments. However, since the discovery of high-pressure induced protein unfolding and denaturation by Nobel laureate P.W. Bridgman in 1914 [3], there is strong evidence that high hydrostatic pressures may lead to disruption of intermolecular interactions maintaining the native protein structure. In addition, thermodynamic parameters do not only influence the protein’s structure and folding process, but also the role of the solvation liquid water. Unfortunately, the experimental conditions do often restrict access to information about the mechanisms behind.

Scientists are also more and more fascinated by new habitats covering as well hot vents in the deep sea where organisms survive at temperatures close to the boiling point of water and high pressure [11], permafrost regions in the Antarctic, or high pressure environments as the inner crust of the Earth. Survival under extreme conditions attracts more attention in a world where viability at normal conditions becomes harder and harder. Temperature variations influence the thermal energy as well as the volume of a system and that makes difficult to separate the two effects. Pressure is another thermodynamic variable that only affects the system volume and thus, the energy variation is better defined from a thermodynamic point of view. Therefore, high hydrostatic pressure is a better variable for exploring molecular dynamics of biological systems under well-defined and controlled thermal conditions, as done recently for instance for the natural nanoparticles lipoproteins [9].

Such considerations have motivated the development of new high pressure and low/high temperature sample sticks at the Institut Laue Langevin (ILL). In the following sections, we describe the design and the operation of these sticks. We then present the very successful results obtained in real experimental conditions.

2. Hardware design

In every top-loading cryostat, the sample bore is anchored to a powerful cold source to reduce the amount of heat brought down to the sample space. There is therefore a cold region above the sample region, even when the sample is regulated at room temperature in the calorimeter of the cryostat. The location of this cold region depends on the nature of the cryostat: at the top of the liquid nitrogen bath of a wet cryostat [4] or in front of the first stage of the cryocooler of a dry cryostat [7]. For this project, the choice of a wet cryostat is preferable as it provides more cooling power. In an Orange cryostat of the ILL [4], the cold point is located about 250 mm below the top flange of the sample bore. The temperature of the stick at this position depends on the sample temperature and varies between 80 and 150 K. When circulating helium in the variable temperature insert (during sample cooling), the cold point can even reach lower temperatures, e.g. 71 K for a cold-valve set to 20 mbar. In such conditions, one cannot inject biological samples, liquids used for hydrostatic pressure transmission, and gases like CO, CO₂ and CH₄ under high pressure [8,10,12].

Several ideas were envisaged to cope with this cold point. Wrapping a Thermocoax around the high-pressure capillary failed: the heat source placed in the sample bore reduced too much the performances of the cryostat. The thermal decoupling of the capillary realised with a secondary vacuum allowed to increase the temperature to 206 K at the cold point, but remained insufficient. So we came to the design presented schematically in Fig. 1 where the top and bottom parts of the sample stick are shown. The basic idea consists in thermalising the high-pressure capillary in a volume filled with about 20 cc of helium exchange gas and to decouple thermally this inner volume from the cold parts of the cryostat with a surrounding vacuum space. To ease the mounting of the sample pressure cell, the 700 MPa capillary is centred on the axis of the stick. During the construction, it is inserted from the bottom and guided through the Wilson seal welded inside the head of the stick. It is then brazed at the bottom so that it can be easily replaced.

Fig. 1.

Schematic view of the high-pressure stick. The 700 MPa capillary is thermalised with exchange gas using a Pt₁₀₀ and a Thermocoax heater. The stick is decoupled from the sample volume with an annular space under vacuum. Optionally, coaxial cables, wires or optical fibres can be brought to the sample space.

The temperature of the exchange gas is regulated with a Pt₁₀₀ sensor and a Thermocoax heater. The sensor is placed at the end of a 0.2 mm thick $\emptyset_{int} 3.6$ capillary welded at the top of the stick and containing the wires, roughly at the height of the cold point. The Thermocoax, whose length of 1060 mm covers most of the height of the stick, is introduced in a 0.25 mm thick $\emptyset_{int} 2$ capillary also welded at the top of the stick. The use of stainless steel capillaries facilitates the installation and the replacement of the sensor and the heater. Oblong holes are performed by laser drilling on the capillary of the Thermocoax to ease the heating (see Fig. 2).

Fig. 2.

Photo taken during the assembly of a sample stick. The bellows are welded to the inner tube and the capillaries are assembled together before insertion. The capillary hosting the Thermocoax has been laser drilled and the Pt₁₀₀ is going to be inserted in the copper block brazed at the bottom of the shortest capillary (shown at the right bottom).

This inner volume is surrounded by an annular space tight to helium that can be evacuated with a secondary pump to decouple thermally the high-pressure capillary volume from the cryostat parts. It can also be filled with about 20 cc of helium when the sample must be cooled down to the lowest temperatures, as explained in the next section.

As for any sample stick, some wiring must be brought down to the sample to measure the temperature. In order to protect this wiring from hazardous manipulations and allow the rotation of the whole stick with a rotating plate fixed at the top of the cryostat, a 0.2 mm thick $\emptyset_{int} 3.6 mm$ capillary is installed in the inner volume to bring the wires to the RhFe thermometer of the sample. This capillary is welded at the top and bottom of the inner space and filled with the exchange gas introduced in the sample bore of the cryostat. Up to four optional capillaries can be added to bring wires or optical fibres extending the capabilities of the stick. All capillaries are distributed around the 700 MPa capillary and maintained in place with a few spacers.

As shown on Fig. 3, the rest of the design is similar to the one of standard sample sticks except that the inner chamber is made from two stainless steel 316L tubes joined by bellows to compensate for the differential expansion of the two coaxial tubes welded to the head of the stick. The top flange of the stick closing the sample bore of the cryostat features a triple O-ring sealing to adjust the sample height and rotate the sample around the axis of the stick (see Fig. 4). As usual, the outer tube holding baffles for screening the radiations and aligning the stick in the sample bore of the cryostat is spring loaded so as to close the calorimeter of the cryostat and reduce the gradient of temperature around the sample. Finally, a 3-way valve connected to the helium bath of the cryostat and a primary pump is mounted on the head to facilitate the flush and the injection of helium in the inner chamber.

Fig. 3.

CAD views of the middle and bottom parts of the high-pressure stick. The capillaries are all gathered in the inner volume. The baffles are welded on a dedicated tube which is pushed against the top flange of the calorimeter of the cryostat with a spring located near the head of the stick (near room temperature). When changing the height or rotating the sample, the baffles do not move inside the cryostat.

Fig. 4.

CAD views of the head of the high-pressure stick. The valves and the connectors are distributed around the head so that the user can remotely rotate the stick in the sample bore of the cryostat.

3. User operation

3.1. Installation

Several versions of the sample stick were produced to fit in cryostats providing different sample bore diameters, typically $\emptyset 49$ , $\emptyset 70$ and $\emptyset 100 mm$ . A few sticks also provide 2 or 4-wire in-situ measurements. The tails of the cryostats being specific to each neutron spectrometer, the length of the sticks was chosen to match the longest tail whilst remaining compatible with the shorter ones, and the Pt₁₀₀ sensors were located at an average height. The high-pressure capillary being heated with exchange gas, the location of the thermometer is not critical.

After attaching the pressure cell, described in more detailed in [14], the stick is inserted into the sample bore of the cryostat being flushed with helium gas (3-way valve connected to the He bath). During this operation, the temperature must be higher than 10 K in the calorimeter of the cryostat. The annular space of the stick is evacuated with a secondary pump and 10 to 20 cc of helium gas are injected into the inner volume after flush. The sample bore of the cryostat is also flushed but then filled with about 200 cc of helium gas.

The temperature regulator of the stick is plugged to a 12-pin Jaeger connector. Following ILL standards, the Pt₁₀₀ sensor is connected with the pins 1-6-3-4 and the 80 Ω Thermocoax heater is wired on pins 7 and 12. The temperature controller is set to regulate the temperature with a maximum current of 400 mA and a maximum power of 12 W. The sample temperature is monitored via a second Jaeger connector, 8 pins, wired to the RhFe sensor inserted in the body of the pressure cell (pins 1-2-3-4).

3.2. Operation

In most cases, the required sample temperature is greater than 100 K. Good settings are obtained with a capillary maintained at 295 K and surrounded by an insulating vacuum of ≈10⁻⁶ mbar. The Fig. 5 shows the temperature variations measured without pressure cell inside the quad-heat-exchanger $\emptyset 70 mm$ cryostat of the TOF instrument IN5 [13] and a standard $\emptyset 70 mm$ cryofurnace. These tests were performed with a 100 m³/h dry helium pump from Adixen [15] and similar results can be obtained with a 40 m³/h oil pump. All coolings were performed with a cold-valve set to 40 mbar. The base temperature was reached with a cold-valve tuned to 2 mbar at 10 K so that the controller delivers about 150 mW heating power at that temperature.

Fig. 5.

Temperature variations recorded upon cooling in a $\emptyset 70 mm$ cryostat and a ∅70 cryofurnace. For clarity, the comparison with a standard stick is shown for the cryostat and the temperature variation of the cold point is shown for the cryofurnace. The cool-down time is twice longer because of the heat brought by the stick but a base temperature of ≈1.8 K is reached after stopping the heating and filling the annular space with ≈20 cc of He.

As expected, the 700 MPa stick brings heat and the cool-down time increases by a factor of ≈2. The base temperature reached with the high-pressure capillary regulated at 295 K is 38 K in the cryostat and 35 K in the cryofurnace. After switching the heater off, the base temperature decreases slightly in both cryostats.

To reach lower temperatures, it is necessary to stop heating the capillary and fill the annular space of the stick with ≈20 cc of exchange gas, for example with a link to the helium bath. That is demonstrated in Fig. 5 where a base temperature of ≈1.8 K is reached in both cryostats. For clarity, only the temperature of the cold point measured for the cryofurnace is shown.

The stick can also be used up to 550 K but of course not with pressure cells made from aluminium alloy or null-matrix TiZr [1] for which the temperature is limited to 100°C. Before warming the sample up, care must be taken to evacuate the annular space and heat the capillary to 295 K to avoid a blockage leading to the explosion of the pressure cell. In case such an accident would happen, the pressure cell would act as pressure relief inside the calorimeter of the cryostat. The worst result will be the destruction of the cell and the calorimeter followed by a sudden warming up of the cryostat.

4. Performances

To validate the equipment, we investigated a biological system in solution that had already been very well characterised and described by Trapp et al. [16,17]. Multi-lamellar vesicles (MLV) formed by the phospholipids 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC) are indeed an important reference system for which structural changes as function of temperature and pressure are known. They are a component of cell membranes that serve among others as filter of the cell with respect to the environment. For the sample preparation, DMPC was purchased from Avanti Polar Lipids (Alabaster, USA) and used without further purification. To produce DMPC MLVs, about 100 mg of lipid powder were hydrated in a desiccator from pure D₂O for two days at 40°C following the protocol of Busch et al. [5]. After having deposited the sample into the high hydrostatic pressure sample holder, additional heavy water was added to achieve a sample with an excess of water to guarantee a homogeneous pressure application.

The experiment was performed on the small momentum transfer diffractometer D16 [6] at ILL by measuring $θ - 2 θ$ scans with incident neutron wavelength $λ = 4.75$ Å. We extracted the d-spacing of the lamellae following the procedure described by Aoun et al. [2] as function of pressure and temperature. DMPC presents a marked phase transition that is found at 297 K at normal pressure conditions and shifted to higher temperatures with increasing pressure [16]. The phase transition is accompanied by a distinct variation in d-spacing and a jump in the lipid dynamics. Therefore, we used the d-spacing as indicator for a passage through the phase transition.

The treatment permitted to extract the $(p, T)$ diagram shown in Fig. 6 and to fit the main phase transition temperature $T_{m}$ as function of pressure. The slope was evaluated to $d T_{m} / d p = (21 \pm 0.5) K / 1000$ bar in very good agreement with values in the literature [17]. The experiment therefore validated the good functioning of the high pressure equipment. Both structural and dynamical characteristics can be obtained and correlated to shed light on the functioning of biological systems under specific conditions.

Fig. 6.

$(p, T)$ phase diagram of DMPC MLVs in excess of D₂O deduced from D16 measurements.

5. Conclusion

We have successfully developed sample sticks for cryostat and cryofurnaces preventing the freeze of fluids injected into 700 MPa pressure cells regulated at temperatures varying between 1.8 and 550 K. Not only these sticks are now available to the neutron community and allow researchers to undertake new experiments at all neutron scattering instruments, but they also feature the capability to perform complementary in-situ measurements (e.g. dielectric spectroscopy) and may be applied to other techniques.

Footnotes

Acknowledgements

This work has been supported by the Agence Nationale de la Recherche (project number ANR-12-ISV5-0002-01 LDLPRESS to J.P.). The authors gratefully acknowledge J. Ollivier who kindly accepted to provide the IN5 cryostat for determining the performances of the first $\emptyset 70 mm$ stick.

References

http://www.ill.eu/sane/equipment/high-pressures/tizr-null-matrix-alloy/.

Aoun,

Pellegrini,

Trapp,

Natali,

Cantù,

Brocca,

Gerelli,

Demé,

Marek Koza,

Johnson and

Peters, Direct comparison of elastic incoherent neutron scattering experiments with molecular dynamics simulations of dmpc phase transitions, Eur. Phys. J. E.39(4) (2016), 48. doi:10.1140/epje/i2016-16048-y.

P.W.

Bridgman, The coagulation of albumen by pressure, J. Bio. Chem.19(4) (1914), 511–512.

Brochier, Top-loading cryostat for neutron scattering, ILL Technical report, p. 74, 1977.

Busch,

Smuda,

L.C.

Pardo and

Unruh, Molecular mechanism of long-range diffusion in phospholipid membranes studied by quasielastic neutron scattering, J. Am. Chem. Soc.132(10) (2010), 3232–3233. doi:10.1021/ja907581s.

Cristiglio,

Giroud,

Didier and

Demé, D16 is back to business: More neutrons, more space, more fun, Neutron News26(3) (2015), 22–24. doi:10.1080/10448632.2015.1057051.

Evans,

R.B.E.

Down,

Keeping,

Kirichek and

Bowden, Cryogen-free low temperature sample environment for neutron scattering based on pulse tube refrigeration, Meas. Sci. Tech.19(3) (2008), 034018. doi:10.1088/0957-0233/19/3/034018.

V.M.

Giordano and

Datchi, Molecular carbon dioxide at high pressure and high temperature, Europhysics Letters77(4) (2007), 46002. doi:10.1209/0295-5075/77/46002.

Golub

et al., Submitted, Sci. Rep. (2016).

10.

R.D.

Goodwin, Carbon monoxide thermophysical properties from 68 to 1000 K at pressures to 100 MPa, J. Phys. Chem. Ref. Data14(4) (1985), 849–885. doi:10.1063/1.555742.

11.

Martinez,

Michoud,

Cario and

Ollivier, High protein flexibility and reduced hydration water dynamics are key pressure adaptive strategies in prokaryotes, Sci. Rep.6 (2016), 32816. doi:10.1038/srep32816.

12.

H.E.

Maynard,

J.S.

Loveday,

Klotz,

C.L.

Bull and

T.C.

Hansen, High-pressure crystallography of methane: A low-temperature neutron diffraction study in the 1–5 gpa range, High Pressure Research29(1) (2009), 125–128. doi:10.1080/08957950802570459.

13.

Ollivier and

Mutka, IN5 cold neutron Time-of-Flight spectrometer, prepared to tackle single crystal spectroscopy, J. Phys. Soc. Japan80(Suppl. B) (2011), SB003. doi:10.1143/JPSJS.80SB.SB003.

14.

Peters,

Trapp,

Hughes,

Rowe,

Demé,

J.-L.

Laborier,

Payre,

J.-P.

Gonzales,

Baudoin,

Belkhier and

Lelièvre-Berna, High hydrostatic pressure equipment for neutron scattering studies of samples in solutions, High Pressure Res.32(1) (2012), 97–102. doi:10.1080/08957959.2011.642376.

15.

Pfeiffer Alcatel A100L, http://www.pfeiffer-vacuum.com.

16.

Trapp,

Marion,

Tehei,

Demé,

Gutberlet and

Peters, High hydrostatic pressure effects investigated by neutron scattering on lipid multilamellar vesicles, Phys. Chem. Chem. Phys.15 (2013), 20951–20956. doi:10.1039/c3cp52762j.

17.

Winter and

W.-C.

Pilgrim, A SANS study of high pressure phase transitions in model biomembranes, Ber. Bunsen-Ges. Phys. Chem.93(6) (1989), 708–717. doi:10.1002/bbpc.19890930611.