There are a number of measurement methods that can be used to investigate hydrogen storage materials and systems. Gravimetric and volumetric methods are the two primary methods and the most robust in terms of depth of analysis; temperature-programmed desorption, differential scanning calorimetry, and thermal gravimetric analysis are also used. All measurement methods quantify a change in a measurable property to indirectly calculate the hydrogen storage capacity of a material. It is important to keep in mind that all methods indirectly calculate hydrogen concentration and each have complications associated with indirect methods. For example, characterizing the hydrogen storage capacity of highly porous media has proven difficult using gravimetric and volumetric methods because both require an understanding of the amount of gas displaced, calculated from the skeletal density of the material, and the corresponding buoyancy (gravimetric) and volume calibration (volumetric) effects on measurements.

This section will summarize these most commonly used measurement methods and provides detailed techniques for measuring hydrogen storage properties. The section also provides a series of detailed consideration that may have a significant impact on the performance and accuracy of a measurement. These range from general sample preparation considerations to issues unique to specific types of measurements and instruments.

One clear consideration is that the measurement equipment should be constructed of materials that are both hydrogen compatible and do not have significant hydrogen permeation. Obviously metal that react with hydrogen at low pressures and temperature such as Titanium, Magnesium, Palladium and Vanadium to name a few should not be used. Common copper tubing and brass components are prone to out-gassing impurities that may have an impact on measurements. 316L stainless steel is commonly used in many hydrogen applications. Teflon and many polymer sealing materials have high hydrogen gas permeation rates and are not generally recommended for these measurement applications.

1. Static and Dynamic Measurements

Hydrogen storage testing can be divided into static and dynamic testing based on whether or not the sorption/desorption reaction is allowed to reach equilibrium at any point during the experiment. In volumetric and gravimetric methods for example, hydrogen gas can be introduced or removed either by aliquots (static) or through a variable flow or pressure regulator (dynamic). In thermally driven methods such as TPD and TGA, the temperature of the sample can be raised in steps (static) or continuously (dynamic).

In static testing, a sample in equilibrium at a certain temperature, pressure and composition is perturbed by a sudden change in pressure (volumetric and gravimetric methods) or temperature (TPD and TGA). The sample is allowed to reach equilibrium at some new temperature, pressure and composition before it is perturbed again. This process continues until the sample is fully sorbed or desorbed. This step-by-step, or equilibrium state-by-equilibrium state process allows static testing to be used to determine the thermodynamic properties of a material by van’t Hoff diagram analysis. Static testing is the most frequently reported testing method in literature when using volumetric and gravimetric methods.

Dynamic measurements are characterized by continuous changes in pressure and temperature and do not allow for equilibrium. Although they are not ideal for investigating purely thermodynamic properties, they can provide information on the kinetic activity of a material. When hydrogen flows at a constant rate into an evacuated hydrogen storage system, the pressure in the system increases linearly until the material begins to sorb. After this point, the pressure in the system is a function of the material’s ability to sorb hydrogen; if the system is to remain at constant pressure across the pressure plateau in Figure 1, the rate of hydrogen flowing into the system must be equal to the rate of hydrogen sorbed by the material.

Figure 1: Comparison of static and dynamic pressure-composition isotherms for LaNi5 and FeTi at 298K. Results are after about 200 test cycles.

Dynamic measurements are more useful for evaluating a material’s hydrogen storage performance during application than static measurements. For example, fuel cells require hydrogen at a specified constant pressure to operate efficiently and safely. In order to supply hydrogen at constant pressure, a regulator is connected between the FC and the hydrogen storage system through which the flow rate varies. Dynamic measurements allow testing of storage materials at flow rates representative of the flow through the pressure regulator in the fuel cell/storage system and the dynamic pressure limits associated with such flow rates. These results are highly dependent on the impact of heat transfer on sorption/desorption kinetics.

2. dC and dP Dosing Methods

In isothermal testing, the sample is charged and discharged by changes in hydrogen pressure in the sample holder. Hydrogen can be added step-by-step in aliquots or continuously through flow or pressure regulators. These techniques are used to provide static and dynamic methods of storage testing. Static and dynamic dosing can also be referred to as dC (differential concentration) and dP (differential pressure) dosing, respectively. In dC dosing, a specific amount of hydrogen calculated from the temperature, pressure and volume of the aliquot is dosed to the sample. In dP dosing, a flow or pressure regulator is used to increase the pressure in the sample cell a specific amount per dose. The terms dC and dP originate from the relationship between the different types of dosing and PCTs; dC dosing steps along the PCT concentration axis and dP dosing steps along the PCT pressure axis.

Volumetric instruments typically take steps in concentration (dC dosing) whereas Gravimetric instruments generally take steps in pressure (dP dosing).

Figure 2 shows the relationship between dC and dP dosing and PCTs for a metal hydride. dC dosing (squares) enables limited investigation of the solid solution phases, represented by the regimes with greater slope at low and high concentrations, compared to dP dosing (circles). However, the reverse is true for the investigation of transitional regimes at intermediate concentrations. dC dosing can provide much greater information across plateaus in metal hydrides and the saturation regimes of physisorbing materials. A common complaint in hydrogen storage testing is incomplete or uninformative investigation of metal hydride plateau regimes due to dP testing. In some cases where the plateau slope is very small, the plateau may be missed altogether.

Figure 2: PCT of classic metal hydride material with plateau. dC (squares) and dP (red circles) dosing provide different information in solid solution (vertical) versus hydride formation (horizontal) portions of the PCT phase diagram.

An advantage of the dC stepping of the volumetric method is that samples can be prepared to specific hydrogen concentrations. This may be very useful for ex-situ examination of the materials using other analytical techniques.

3. Volumetric Method

The volumetric method of hydrogen storage measurement, also known as the manometric method, or as Sievert’s method in honor of the German chemist of the same name, uses temperature-pressure-volume correlations to determine hydrogen concentration and the storage properties of a material. A generalized volumetric system with commonly employed components is shown in Figure 3.

Figure 3: Multipurpose Gas Sorption/Desorption Apparatus. The pressure gauge on the right is only required when using the flow controller.

The apparatus consists of a gas reservoir connected to a specimen reactor. Because the volumetric method measures concentration indirectly through temperature-pressure-volume correlations, the volumes and temperatures of the reservoirs and sample holder (system) must be known in advance. A thermocouple should always be positioned in as close a contact as possible with the sample. The system volumes are carefully pre-calibrated and the reservoir and sample holder are maintained at constant (but not necessarily equal) temperatures using an external temperature controller. By fixing volume and temperature, reservoir and sample holder pressures can be measured using pressure transducers to provide isothermal pressure-concentration data.

The volume of the sample holder must be calibrated while filled with sample in order to get an accurate measurement of the free gas volume in the system. Although a generally straightforward procedure for chemisorbing media, calibration can be tricky for highly porous and nano-structured media at low temperatures due to inaccurate or incomplete skeletal density information and the possible physisorption of helium used to perform the calibration. Note however, that it is unlikely that most materials will have significant helium adsorption below 5 bar and at 77K, and certainly not at significantly higher temperatures. It can be verified that helium does not adsorb by measuring the sample holder volume with the sample (free volume) as function of the helium expansion pressure at temperature. Materials that do not adsorb helium will likely show a constant free volume as function of pressure. Similarly, in a gravimetric measurement, such material would also exhibit a linearly decreasing mass response (buoyancy) as function of helium pressure.

Volumetric testing requires accurate measurement and control of the instrument and sample holder temperatures and the associated temperature gradients. The temperature of the instrument should be controlled in order to minimize temperature fluctuations due to external sources such as room heating, ventilation and air-conditioning. The temperature of the sample can be precisely controlled below room temperature by cryogenic cooling accompanied by PID (Proportional Integral Differential control) heating or above room temperature by PID heating and appropriate insulation.

To test a sample using the static method, the gas reservoir is filled with H2 to a pressure P and then allowed to react with the specimen reactor by opening the top valve. When filling the gas reservoir, sufficient time should be allowed for the gas to come to thermal equilibrium with the tubing and vessels before a pressure reading is taken. An equilibrium pressure P’ is reached between the gas reservoir and the specimen reactor that, once paired with temperature, initial pressure and volume information, can be used to precisely determine the amount of hydrogen sorbed by the material. Dynamic testing allows the slow and continuous introduction of hydrogen to the specimen reactor. Hydrogen gas from the gas reservoir flows into the specimen reactor via the electronic flow controller while continuous pressure readings are taken by the pressure gauges attached to the reservoir and reactor. The pressure, temperature and volume data are analyzed to generate dynamic (non-equilibrium) pressure-composition isotherms.

One intrinsic advantage of the volumetric method is that, unlike gravimetric methods, the quantity of gas dosed to or from the sample can be small compared to the total capacity of the sample. Thus, the method allows the direct preparation of sample of known hydrogen contents. This may be useful for doing in-situ or ex-situ secondary measurements such as X-ray or neutron diffraction, NRM, IR, Raman, etc.

4. Gravimetric Method

The gravimetric method of measurement uses weight changes measured on a balance to determine concentration and the storage properties of a material. A schematic of a simple symmetrical microbalance gravimetric system is presented in Figure 4. Note additionally, that gravimetric systems are also often configured as flow-through systems, in which case there is also a gas exit port attached to the chamber through a pressure control device.

Figure 4: Schematic of counterbalanced gravimetric method system for hydrogen storage testing

Before gravimetric testing, the weight of the sample is measured and the sample is placed on one end of a symmetric microbalance in a sample holder. In symmetrical microbalance gravimetric instruments, an inert tare with the same weight and comparable density to the sample is placed on the other end of the microbalance to provide a counterbalance. The tare is designed to minimize the effects of buoyancy caused by the hydrogen gas displaced by the sample volume and must be inert in a hydrogen atmosphere. Note that at relatively high pressures, buoyancy measurements should be considered as an intrinsic part of the gravimetric method, basically on the same standing and care as the free (dead) space volume calibration in the volumetric method.

After the chamber containing the gravimetric equipment is evacuated, hydrogen from an external hydrogen source enters the chamber and is generally increase in incremental pressure steps as it is sorbed by the sample in the sample holder. The microbalance is typically equipped with an electronic circuit that measures the strain on the balance material that is directly related to the change in weight of the sample. This information combined with pressure and temperature readings provided by a pressure gauge connected to the chamber and a thermocouple located next to the sample holder are needed to measure hydrogen storage properties.

Measurements can generally be made under isobaric conditions with no loss of sensitivity and modification of the thermodynamic driving force or altering the kinetics by significant pressure changes. This can be accomplished even in static experiments by providing makeup gas during absorption or bleeding the evolved gas during desorption. Finally, special care must be taken to avoid contamination gases in the hydrogen using this method.

5. Thermal Gravimetric Analysis Method

We make a distinction here between the Gravimetric instruments described above and Thermal Gravimetric Analysis (TGA) in that Gravimetric instruments generally operate under isothermal conditions with a controlled over pressure of gas. Whereas, TGAs typically operate under vacuum or low-pressure flowing gas conditions and ramping temperatures. Some equipment, however, may be setup to operate in all of these modes. TGA is a thermal analysis technique often used in conjunction with DSC to determine the hydrogen storage properties of a sample. The first documented use of TGA was in the study of the efflorescence of hydrated salts in 1912 but it is now traditionally used in the quantitative investigation of decomposition reactions. TGA is experimentally similar to DSC but instead of measuring the heat flow as function of temperature, TGA measures sample weight as a function of time or temperature. This is accomplished by placing the sample in an environment that is heated or cooled at a controlled rate and measuring the weight change.

The equipment necessary to perform TGA consists of an accurate balance, a programmable furnace, a reaction chamber and a data collection system. As the temperature inside the furnace and reaction chamber changes, the balance measures the variation in weight due to various chemical reactions including dehydriding. TGA is affected by many of the same issues as the gravimetric testing method such as buoyancy and mechanical disturbances. The buoyancy force exerted on the sample by the displaced fluid varies with temperature and must be taken into account either during the experiment, e.g. the tare technique in the gravimetric method, or during data analysis; mechanical disturbances must be minimized through leveling and anti-vibration supports. Another complication TGA shares with the gravimetric method is how to determine the sample temperature. Using a thermocouple to directly measure the temperature affects the weight reading of the sample. Consequently, care must be taken to place the thermocouple in such a way that it accurately reads the sample temperature but does not affect weight readings.

Thermal gravimetric instruments require calibration on both temperature and mass. Temperature calibration for DTA-TG (Differential Thermal Analysis – Thermal Gravimetric) instruments is performed using the extrapolated onset temperature of the DTA peak of the melting points of standard materials. For TG instruments, temperature is calibrated by the Curie point method or the standard wire melting method.

Mass calibration is performed at room temperature by using standard masses—removal of a standard weight on the sample holder, where the recorded change in mass is calibrated to the mass of the standard weight. The drift in baseline upon heating should be checked by a measurement using a blank sample pan. Following the measurement with a blank, a standard sample which decomposes quantitatively by several well-separated reaction steps is subjected to the same measurement (i.e., heating rate, sweep or carrier gas flow rate and type, final temperature). The corrected mass change is constructed by subtracting the apparent mass change as a function of temperature from the sample under study. The experimentally corrected mass loss should be confirmed and in good agreement with the calculated mass loss from the reaction stoichiometry.

As with DSC (Differential Scanning Calorimetry) measurements, the sample (solid or liquid) is weighed into a small crucible/sample pan. Typical sample pans are made of carbon, aluminum, silica, platinum, stainless steel, or inconel. The sample pan is selected based on material compatibility with the sample and products. Sample size should be kept to a minimum, but large enough to observe the necessary mass changes. With respect to solid samples, the particle size is a critical parameter that will influence the shape and position of the TG curve. Typical TG experiments are performed under either a constant heating ramp or under isothermal conditions. In the isothermal runs, mass change is recorded as a function of time while at a constant temperature. For the case of non-isothermal runs, the mass change of the sample is recorded as a function of temperature while being subjected to a prescribed linear heating rate. A means to reduce the temperature gradients present within the sample is to impose slower heating rates (typical heating rates rarely exceed 10 k/min).

Scanned temperature experiments (TGA) are often done using flowing inert gas, such as argon. Although useful for comparing different materials (catalyzed vs. uncatalyzed, for example), desorption occurs into an essentially H2-free environment, and thus it does not reflect the actual conditions encountered in a real storage system application, where there is always H2 gas present at pressures at or exceeding 1 bar. Interpretation of desorption temperatures thus must be done carefully. For this reason TGA may be used as a screening tool, but not as a technique to rely on for realistic system desorption temperatures. When one really wants to know practical desorption temperatures, the experiment must be conducted by flowing H2 gas into the TGA at the particular pressure of interest (say, 1-8 bar).

As mentioned above, in most cases the TG experiments are performed under either an inert carrier gas or a reactive gas. The flow rates and type of gas will affect the apparent mass change. Consequently, buoyancy calibration runs need to be made with an empty pan for a given gas composition, heating rate, and purge/sweep gas flow rate. Buoyancy calibrations need to be performed on a regular basis, ideally every time a sample is performed.

6. Temperature-Programmed Desorption Method

Temperature-Programmed Desorption (TPD) refers to a wide range of experimental methods that rely on temperature variation and generally include mass spectroscopy to investigate and quantify desorption reactions. The technique can be used for both reversible and irreversible processes, with the latter referred to as Temperature-Programmed Reaction Spectroscopy (TPRS).

TPD measurements may be as simple as measuring pressure rise in a volumetric instrument or weight loss in a gravimetric instrument while ramping the sample temperature. However, TPD generally refers to spectroscopic desorption measurements. The basic set-up for TPD techniques is illustrated in Figure 5. The sample is loaded into the experimental apparatus (a temperature-programmed heater contained in a vacuum chamber) and charged with hydrogen until fully loaded. Note that hydrogen loading may also be performed in a separate apparatus prior to putting the sample in the TPD instrument. After the remaining gas has been drawn off, the computer-controlled heater slowly raises the temperature of the sample. This releases hydrogen that is evacuated to vacuum. A mass spectrometer connected to the evacuation line analyzes the relative composition of the desorbed gas and quantifies the amount of hydrogen desorbed by the sample. As with other methods (gravimetric and volumetric) that employ simultaneous mass spectrometry analysis, TPD systems have an advantage, in that it can distinguish between hydrogen and other constituents in the evacuated stream.

Figure 5: Schematic of typical Temperature-Programmed Desorption experimental set-up

Concentration, temperature and time data is obtained through TPD measurements and can be used to determine capacity, kinetics and thermodynamic properties of a material. A common data representation for TPD experiments is shown in Figure 6.

Figure 6: Typical data representation of single-component TPD experiment with CO.

Figure 7: a testing report of TPD curve by H-Sorb 2600 high pressure adsorption analyzer

The area under the peak at ~475K in Figure 6 is proportional to the amount originally sorbed to the sample. In the case of full charging, it represents the capacity of the material. Kinetics information is obtained from the contour of the spectroscopic peak and knowledge of the relationship between temperature and time based on the computer program. Lastly, the temperature associated with the peak is related to the reaction enthalpy of hydrogen-substrate desorption. Unfortunately, TPD measurements can only be done at vacuum due to limitations in analytical equipment, effectively limiting the amount of thermodynamic information that can be collected. Quantitative analysis requires accurate calibration of the mass spectrometer against known flow rates. It is also important to understand that TPD or any dynamic type measurement of measurement gives results that are a convolution of both thermodynamic and kinetic properties of the hydrogen storage material. This may lead to miss-interpretation of the results. For example, does a decrease in desorption temperature of a material that is modified through the addition of a dopant mean that the thermodynamics of the host material has been altered? Or, is the dopant acting as a catalyst, or a thermal conductor, or modified the materials morphology such that kinetics are improved, increasing desorption rates at lower temperatures? For the development of new (reversible) materials it is important to have information on both the dynamic and equilibrium hydrogen sorption behavior of the storage material.

7. Differential Scanning Calorimetry Method

Differential Scanning Calorimetry (DSC) is a thermal analysis technique used to investigate the thermodynamic properties of a material by measuring the energy necessary to maintain a sample material and an inert reference material at the same temperature over a range of temperatures. The relative heat flow to the sample material as a function of temperature can be used to determine thermodynamic properties such as specific heat and enthalpy.

Temperature variation in DSC is controlled by a computer and is typically linear in order to simplify calculation, although nonlinear temperature variation can be used as well. The specific heat at constant pressure Cp of a sample material (on a per mass unit basis) as a function of temperature is determined by the equation:

where T is temperature, h is energy, q is heat flow and t is time. The heat flow can be taken from the data and the temporal temperature variation is based on the computer program controlling the experiment. In this way, the specific heat as a function temperature can be determined. Furthermore, the enthalpies of reactions can be determined from the measurement. 

The two most commonly used methods for conducting DSC measurements are power-compensation DSC and heat-flux DSC. In the power-compensation method, the sample and reference material are placed in independent, identical furnaces. The furnaces are maintained at the same temperature over a variety of temperatures by varying the power input. The power input and temperature data are used to construct the DSC diagram. The indirect and direct variables in power-compensation DSC are flipped in heat-flux DSC. The sample and the reference material are placed in one furnace and exposed to the same heat flux. The variation in temperature between the sample and reference is used to determine the relationship between heat flux and temperature.

In the context of hydrogen storage, Differential Scanning Calorimetry is primarily used for desorption testing because DSC equipment is not typically designed to handle the high pressures required for sorption with some exceptions. The advantage of DSC over other methods is that other thermal events such as melting or crystal structure changes may be observed. A significant limitation for testing hydrogen storage using DSC is that there is no way to determine the amount of hydrogen desorbed by a sample, only the total enthalpy of a given reaction. For instance, DSC alone would be unable to distinguish between materials that desorb 0.1 mol H2 with an enthalpy of reaction of 30 kJ/mol H2 from one that desorbs 1 mol H2 with an enthalpy of reaction of 3 kJ/mol H2. It is important to have an understanding of the enthalpy of reaction per mol hydrogen in order to compare thermodynamics across materials. This vital consideration for hydrogen storage should not be overlooked but can be remedied by coupling DSC with quantitative measurements of hydrogen uptake and release.

8. Differential Thermal Analysis Method

Differential thermal analysis (or DTA) is a thermoanalytical technique, similar to differential scanning calorimetry. In DTA, the material under study and an inert reference are made to undergo identical thermal cycles, while recording any temperature difference between sample and reference. This differential temperature is then plotted against time, or against temperature (DTA curve or thermogram). Changes in the sample, either exothermic or endothermic, can be detected relative to the inert reference. Thus, a DTA curve provides data on the transformations that have occurred, such as glass transitions, crystallization, melting and sublimation. The area under a DTA peak is the enthalpy change and is not affected by the heat capacity of the sample.

A DTA consists of a sample holder comprising thermocouples, sample containers and a ceramic or metallic block; a furnace; a temperature controller; and a recording system. The key feature is the existence of two thermocouples. One thermocouple is placed in an inert material such as Al2O3, while the other is placed in a sample of the material under study. As the temperature is increased, there will be a difference in the temperatures (voltages) of the two thermocouples if the sample is undergoing a phase transition. This occurs because the input of heat will raise the temperature of the inert substance, but be incorporated as latent heat in the material changing phase.

Today most commercial instruments are no longer true DTA devices but rather have incorporated this technology into a Thermogravimetric analysis equipment (TGA), which provides both mass loss and thermal information. Even these instruments are being replaced by true TGA-DSC instruments that can provide the temperature and heat flow of the sample, simultaneously with mass loss.