Plasma catalysis: a brief tutorial

Plasma comprise 99.9% of the visible universe. The stars including the sun, are examples of plasma. Plasma can be simply defined as a quasi-neutral gas containing charged and neutral particles which exhibit collective behavior [1]. Plasma contains electrons, neutrals, electronically and vibrationally excited species, ions, radicals and atoms. Plasmas can be broadly classified into two types: (1) Thermal, (2) Non-thermal. Thermal plasmas are in thermodynamic equilibrium, the electron temperatures as well as the ion temperatures are the same, which increase the bulk temperature between a range of 3000–5000 K kelvin without external heating. This type of plasmas is generated at high pressure with high excitation energy. Flame and fusion plasmas are examples of thermal plasmas. Majority of hot plasmas are found in space whereas earthly application of thermal plasmas is limited to astronomy research, fusion reactors, particle accelerators and other particle physics research. The higher electron temperatures in thermal plasmas make them more suitable for decomposition reactions which led to its application in forming metal nanoparticles [2] and plastic waste treatment [3] where breaking of bonds is quintessential. Thermal plasmas in the low temperature spectrum have also found application in metal processing due to the high melting and boiling point of majority of metals.

On the other side, the so called 'Non-Thermal Plasma' has low enthalpy which makes its study relatively simpler and its applications much wider as compared to thermal plasmas. Furthermore, the bulk gas is typically at room temperature whereas the electrons are at temperatures in the order of 10⁵ K, due to their small mass. Non-thermal plasmas have found extensive applications due to their low bulk gas temperatures and operation at low and atmospheric pressures. Non-thermal plasmas are also known as cold plasmas as the electrons are much hotter than the ions. A Non-Thermal plasma is also termed as a non-equilibrium plasma as the electron temperatures are much hotter than the ion temperatures. Typically, the electron temperature in non-thermal plasmas range from 0.01–16 eV. This range is extremely suitable for chemical reactions as the bond dissociation and ionization energies of atoms and molecules fall in this regime. Non-thermal plasmas, containing highly energized species can help in discovering alternative routes for several reactions to occur such as ozone synthesis, ammonia synthesis, hydrogen production and methane reforming. Plasma assisted reactions help in overcoming the bond association energy and sometimes even the activation energy barrier required for a reaction to occur. The challenge is to employ successfully targeted excited species in order to form the desired products. This can be achieved with the help of a catalyst. A catalyst can be defined as a material that promotes the breaking and/or formation of definite bonds exhibiting selectivity towards specific compounds.

Plasma catalysis is an approach which brings both elements together, the beneficial effects of plasma with the possible enhanced selectivity of a catalyst. The plasma state can provide high energy species helping the reactants reach either their dissociation energy or activation energy or both in some cases and the catalyst can help in the synthesis of the targeted product(s) due to its possible enhanced selectivity when selected properly. Moreover, plasma can also enhance the catalyst surface by removing impurities, improving the dispersion of the active phase on the surface of the catalyst and reducing coking, for example. Even though the efficiency of plasma-catalyst synergy is proven for several reactions such as NO_x treatment, ammonia synthesis [4], methane reforming [5], hydrogen synthesis [6], carbon dioxide fixation [7, 8] and ozone synthesis [9], a clear understanding of the underlying phenomena when a catalyst is exposed to plasma is necessary to further advance this technique. This was demonstrated by the commercialization of the plasma synthesis of ozone, where a detailed understanding of the ozone synthesis mechanisms was essential for its commercial scale application. Some important plasma applications are summarized in figure 1.

Zoom In Zoom Out Reset image size

Before providing details about the different plasma discharges, it is important to define some useful terms which are used frequently in plasma catalysis and that the reader might found in the following sections. A catalyst is a material that helps to reduce the activation energy required to convert reactants to products by forming intermediates. Electron temperature is the energy of free electrons in the plasma, usually expressed in electron volts (eV) or kelvin (K). Ion temperature is the energy of the ion or excited species depending on the chemical formula of the specie, usually denoted in eV or K. Electron density is the number of free electrons per unit volume of plasma. Electron energy distribution function (EEDF) f (ε), is the probability density for an electron to have energy ε. Electronically Excited species are the atomic or molecular species having its electrons in higher orbitals as compared to its ground state. Ions are defined as the charged species in plasma other than electrons. Neutrals are unexcited atoms or molecules in the ground state. Atomic species are species having only one atom in its formula and are electrically neutral, while they may or may not be excited. For the ocurrence of atomic species, the molecule goes under electron impact dissociation while for vibrational excited species the electron transfers its energy to the molecule increasing its vibrational temperature, allowing the electrons to jump between the hybridized orbitals with higher energy states. The species adsorbed on the surface of the catalyst or wall of the chamber (irrespective of the charge or the excitation level) are defined as surface adsorbed species. Ion sheath layer is defined as a thin region (a sort of skin) surrounding the plasma. Sheaths act to balance the electron and ion currents lost from a plasma. The most common type of sheath is an ion sheath, meaning it contains more ions than electrons. Cross section can be understood as an imaginary circle of area σ, moving together with one of the collision partners. If the center of the collision partner crosses the circle, then the elementary reaction takes place. The words, plasma and discharge are interchangeable in this tutorial.

To understand plasma-catalysis, it is necessary to review some basics of plasma. Starting with the IUPAC definition of plasma as follows: 'A gas which is at least partially ionized and contains various types of particles, viz. electrons, atoms, ions and molecules. Plasma as a whole is electrically neutral'. In other words, a gas confined in an electrically isolated space having ionized and excited species can be called plasma. Overtime with the discovery of new plasma sources, this definition has evolved replacing the term 'electrically neutral' with 'electrically quasi-neutral'. Primarily non-thermal plasmas (NTP) can be classified based on the electron temperature (see table 1). Important examples of NTP employed for chemical processing include: (1) Dielectric barrier discharge, (2) Radiofrequency discharge, (3) Microwave discharge, (4) Corona discharge and (5) Gliding arc discharge. These plasmas can be used for plasma-catalytic reactions but the criteria to choose the type of plasma varies extensively from case to case. We will focus the discussion in this section on these five discharges. It is important to mention that most plasmas of practical significance have electron temperatures between 1–20 eV, with electron densities in the range 10⁶–10¹⁸ cm^–3. Remember that high temperatures are conventionally expressed in electron volts; being 1 eV $\cong$ 11 600 K.

The general classification of plasmas is summarized in table 1. (Modified from Patil et al [10]).

2.1. Plasma Catalysis time evolution

Plasma itself is a high-tech tool that allows to activate reactant molecules achieving rates of reactions, yields, conversions and selectivity otherwise inaccessible by thermal catalysis at the same conditions. As Lieberman and Lichtenberg stated in their Principles of Plasma Discharges and Materials Processing book: plasma processing is born out of the need to access a parameter space in materials processing unattainable by strictly chemical methods .

Plasma-catalysis, which is the combination of plasma with a proper catalytic material for such active environment, is an emerging and highly promising branch of plasma processing. Plasma catalysis is achieved by placing a catalyst in plasma or after-plasma region of the discharge leading to an improved processing of the input gas stream. In the past few decades, this technology has been applied to a number of gas-phase reactions such as hydrocarbon reforming [51], pyrolysis [52], hydrogen production [53], NO_x decomposition [54] and ammonia synthesis [4] to name a few. However, plasma catalysis is not limited to the synthesis of important chemicals of great societal interest. This technique can be extended to the synthesis of materials, where a catalyst is placed in the discharge zone in order to make possible new growth mechanism to occur. Hence, to clearly understand the plasma catalytic process, it is important to look at this technique from two main different perspectives. From materials processing point of view, plasma has the potential of improving the dispersion of supported metals [55], leading to the possible increase of active sites number. Moreover, improved BET surface areas have been reported with plasma treatment of catalysts when employing gas sources such as nitrogen, argon and ammonia [56]. Furthermore, creating and healing catalyst defects with plasma is also a possibility and this techniques offers also enhanced control over these processes [57]. Plasma treated catalysts have also shown greater impunity towards poisoning [55]. These beneficial effects have been achieved by pre-reaction treatment of the catalysts and serve as further motivation to explore in detail the catalyst synergy with plasma. However, a detailed understanding of the mechanisms that make possible the synthesis of these 'enhanced-plasma catalysts' is not well understood yet. Hence, a detailed knowledge can help us not only to have a better control of the desired outcomes, but it can help to find important information that might be employed for the synthesis of chemicals when using plasma catalysis.

It is important to mention that plasma itself has its own positive outcomes. Among them is the higher concentration of excited and ionized species as compared to conventional thermally heated gas-phases. The presence of plasma also ensures constant delivery of excited species to the catalyst surface. Thus, it can help to achieve higher probability of fruitful collisions. Moreover, on collision with a solid, even with the reactor wall, plasma creates hot spots yielding to higher catalytic activity. In some cases, the reactor wall itself has shown to have a catalytic effect, which is the so called 'wall-effect'. Main functional applications of plasma catalysis for synthesis of specialized materials and for specialty chemical production is shown in figure 2.

Zoom In Zoom Out Reset image size

In general, plasma-catalysis can be performed in two main different arrangements: (a) One stage plasma catalysis, sometimes also called in-plasma catalysis, where the catalyst is placed in the glow region (figure 3(a)). And two stage plasma catalysis, also called sometimes post-plasma catalysis, where the catalyst is placed in the post-glow region (figure 3(b)). Other arrangements exist and can be as complex and varied as the application requires. And the best method to choose depends on the desired outcome. It should be noted at this point that most of the researchers working on plasma catalysis employ their own reactor designs that are adapted to the needs of the reaction/process under study. Hence, this is just a very broad classification and as it can be inferred, thedesign of plasma reactors is an important engineering task to perform with the outmost care by plasma-catalysis researchers.

Zoom In Zoom Out Reset image size

Moreover, with the ever-improving efficiency of power supplies, plasma systems are becoming more energy friendly with the possibility of allowing operation from renewable electricity sources. This potential recourse has been driven by the prospect of inexpensive renewable electricity—that would also reduce the carbon footprint of the process—as prices continue to fall (e.g. commercial solar power fell from $5.36/watt in 2010 to ≈$1.85/watt in 2017 [58]).

The reactors used in plasma catalysis are classified based on the type of non-thermal plasma discharge employed. In all the cases, a strong electro(magnetic) field is used to excite the working gas. The following sub-sections present general schematics and descriptions of the widely used configurations for plasma-catalysis. Many more customized configurations can be found in literature. Due to this wide variety of reactor's design it is important to introduce the reader to generalities related to these discharges.

4.1. Generalities of the glow discharge

The glow discharge is probably the most widely known type of non-thermal plasma discharge. It has been used in plasma chemistry processes for more than a century. And as its name indicates this discharge is luminous. This type of discharge can be defined as a self-sustained continuous DC discharge with cold cathode, which emits electrons as a result of secondary emission mostly induced by positive ions [59].

Glow discharges are generated in vacuum. Traditionally, two electrodes are sealed in a quartz tube. The electrodes are connected to a DC power supply while there is vacuum inside the tube. Both the electrodes are typically in-contact with the gas. The DC power supply can also be changed to AC power supply, but the frequency must be at least 500 Hz to drive the plasma. A representative schematic is shown in figure 4. More recently, these discharges can also be operated at atmospheric pressure at low frequencies (<100 kHz) by connecting it to a high voltage AC power supply. Some common applications of glow discharges are to produce atmospheric pressure non-thermal plasma torches [60], spectroscopy [61] and neon signs [62].

Zoom In Zoom Out Reset image size

4.2. Radiofrequency discharge reactors

As its name suggests radiofrequency (RF) waves are used to excite the working gas. The common operational frequency is between 500 kHz–300 MHz. The voltage to current ratio is between 10⁰ to 10². Two types of reactor setups can be employed with this discharge viz. capacitive (RF CCP) and inductive (RF ICP). The capacitive setup (CCP) cannot be operated at atmospheric pressure, since an arc discharge can be formed. In contrast the ICP setup can be operated at atmospheric discharge using an RF power supply. To avoid interference with radio communication systems, specific frequencies have been assigned for operation of industrial RF discharges; being 13.56 MHz and 27 MHz the most commonly used. Even though both the discharges are excited by similar energy sources their characteristics can vary significantly. For both setups typically the plasma chamber is made of stainless steel or quartz. The chamber is surrounded by RF coils or electrodes. In inductive setup the RF coils are usually outside the chamber while in capacitive setup the electrodes are in contact with the working gas. Inductively coupled reactors are commonly employed for chemical synthesis [63] or specialized material fabrication such as carbon nanostructures [64] or semiconductor growth [65–67] whereas capacitively coupled reactors are used for plasma torches [68], detectors [69], atomizers [70], etc. These plasmas often require complex power supplies nedeed to deliver optimal power to the working gas. The ICP discharge has lower electric fields than capacitively coupled plasma (CCP) discharge. And it is due to this that the ICP discharges at moderate to high pressures usually generate thermal plasma, whereas the RFCCP discharges at the same conditions can generate non-thermal plasma. These discharges can reach temperatures of few hundred kelvins without external heating and they often require water cooling. A schematic of both basic configurations is depicted in figure 5.

Zoom In Zoom Out Reset image size

4.3. Microwave discharge reactors

Microwave (MW) discharges are considered similar to radiofrequency discharges. However, in contrast to RF discharges, microwave plasma is sustained by centimeter range electromagnetic waves. Thermal plasma generation in microwave is usually related to high-pressure systems. Microwave generators, specifically magnetrons, operating with power exceeding 1 kW in the giga hertz frequency range, are capable to maintain the steady-state thermal microwave discharges at atmospheric pressure. Electromagnetic energy in microwave discharges can be coupled with plasma in different ways. The most typical coupling is provided in the form of waveguides, where a dielectric transparent tube for the electromagnetic waves (usually quartz) crosses a rectangular waveguide. This type of discharge can operate at atmospheric pressure or in vacuum. It operates on the same principle as a microwave oven, but the setup varies significantly. The operational frequency is between 300 MHz to 300 GHz but most power supplies are available with a frequency of 2.45 GHz. The setup requires constant cooling of the plasma chamber and as mentioned before it employs typically the use of a square (or rectangular) waveguide for optimal power delivery. A three-stub tuner is used to tune the impedance of the reaction chamber. MW plasma reactors can reach up to 5000 K [71] which is why constant cooling of the chamber is required. These reactors are suitable for torches employed in synthesis of metal [72–74] and ceramic nanoparticles [75, 76] . Another application is waste decomposition [77, 78] and methane pyrolysis [79]. A schematic of a typical Microwave plasma reactor is shown in figure 6.

Zoom In Zoom Out Reset image size

4.4. Dielectric barrier discharge reactor

In 1857 Ernst Werner von Siemens discovered the silent discharges, today known as 'Dielectric Barrier Discharges' [80]. This discharge is produced by applying high voltage to one of the two electrodes separated by a dielectric material. Over the years, there have been many modifications, and the definition has broadened. Currently, discharges that are operated in the frequency of 0.05–500 kHz can be considered as dielectric barrier discharges. The electrodes are usually made of either a metal or graphite whereas the dielectric material is chosen to be quartz, glass, alumina, enamel, mica, teflon, silicon rubber and even some plastics. The applied voltage ranges from 1–100 kV_rms [81]. The DBDs are widely used for pollution control and exhaust cleaning from CO [82], NO_x [83], SO₂ [84] and VOCs [85], biomedical applications, lasers. This discharge most widely known application is plasma television [59].

DBDs reactors work at atmospheric pressure, which makes them very attractive for industrial applications. The plasma in this type of discharge can be classified as non-thermal in nature which sometimes is not the case when RF and MW discharges are run at atmospheric pressure as explained in previous sections. In other words, the electron temperature as well as the gas temperature in an atmospheric pressure DBD is lower as compared to electron temperature in atmospheric pressure RF and MW discharges. This attribute of DBD was an important factor for the successful commercialization of plasma catalytic ozone reactors and air purification systems. The discharge in a DBD is limited between the electrodes. The microdischarges a.k.a. streamers formed in the discharge zone also help to create a better interaction between the catalyst and the plasma species. The microdischarges interacting with a catalyst in a DBD make the electric field non-uniform and stronger by a factor of 10–250 which makes the plasma achieve a non-equilibrium state. Furthermore, in a DBD the unionized species sheath layer is just few micrometers thick..

Dielectric barrier discharge (DBD) is considered the most versatile setup for chemical synthesis. It is usually operated at room temperature and atmospheric pressure. There are innumerous designs in literature. The electrode can be made of any conducting materials while the plasma characteristics depend on the discharge gap. This reactor is extremely versatile however, the discharge gap must be designed with extreme precaution as it governs the formation of glow discharge or an arc. One of the most widely used appliances of this technology are plasma televisions. In a DBD, the electrodes can be arranged coaxially, concentrically, on top of each in case of plates (separated by a dielectric) and recently, floating electrode design has also been employed. The configurations of various DBD setups are shown in figure 7. It should be noted that for plasma-catalysis the concentric arrangement is preferred (figure 7(a)). One of the electrodes is chosen to be the high-voltage electrode and the other one acts as a ground. The current is usually in the range of micro- to milli-amperes. As the ratio approaches 1, the discharge tends to become more unstable and arcs between the two electrodes. Catalysts can also affect the electric field, as a result typically insulators are employed as catalysts. The power consumption is extremely low as compared to other glow discharges. It should be noted that the nature of the packing is very important in any plasma-catalytic reactor. In fact, the most used type of DBD for plasma catalysis, that uses packing material, is the so-called packed-bed DBD reactor. And as an example, we can point out the case when a DBD reactor is packed with a dielectric material such as BaTiO₃. This latest can be considered as ferroelectric and polarization effects can lead to enhanced electric fields at the contact points between the grain boundaries resulting in a higher concentration of excited species. Hence the nature of the discharge changes, from streamer propagating in the gas phase (no packing) to a mixture of gaseous and surface discharge (with packing) a widely used discharge due to its versatility and atmospheric pressure operation conditions. As previously discussed, DBD discharges can be arranged in various configurations which enhances the applications for this versatile discharge. One widely employed geometry is the concentric DBD reactor. In this configuration, the high voltage electrode is a metallic rod, copper for example, with a smaller diameter than a quartz tube where this rod is placed inside which is itself the reactor wall. A gap between the high voltage electrode and the quartz tube allows the flow of gases and plasma generation. The second electrode or outer electrode also known as the ground electrode can be a metal mesh wrapped around the outside of the quartz tube, metals such as copper and stainless steel can be used for such purpose. The length of this mesh determines the plasma zone length. The catalyst is typically packed in the gap between the inner electrode and the quartz tube. A high voltage power supply is used to apply a voltage in the range of 1–30 kV while the current is less than 1 mA. Optimum power is delivered by tuning the frequency of the electric signals.

Zoom In Zoom Out Reset image size

When using a dielectric barrier discharge one usually gets a streamer discharge which is a result of electron avalanches. With the introduction of streamers, the localized electric field momentarily increases the electron field thereby increasing the probability of more molecules getting vibrationally excited as well as reaching higher levels. But in-order to form these streamers, the reactor must meet meek criterion. Meek criterion suggests that in order to form streamers in the void space between the two electrodes the value of αd must exceed 20. Where α is the ionization coefficient and d is the discharge gap. In other words, the electron density must be higher than 3 × 10⁸ cm⁻³ (as $\exp (\alpha d)={n}_{e}$) [86].

Furthermore, plasma emits a varied range of electromagnetic radiation especially in the ultraviolet, visible and near infrared region. And they can play an important role in supplying seed electrons for generating streamers in dielectric barrier discharge reactors.

Several studies have been conducted using DBD as the model reactors. This type of discharge is the industrial benchmark technology for ozone synthesis. Moreover, plasma-catalysis as well as process intensification can be performed in this type of reactors with the catalyst packed in the discharge gap. Besides plasma catalysis, DBDs are used for cancer treatment [87, 88], microbes-inactivation [89], plasma actuators [90–92] and dentistry [93] to name a few.

An important DBD configuration for plasma catalysis is the Surface DBD plasma reactor. The most basic configuration for this type of reactor consists of two electrodes asymmetrically positioned on both sides of a dielectric (barrier) material. Where one of the electrodes is exposed to the surrounding gas and the second electrode can be either buried in the dielectric barrier or covered by a layer of dielectric material such as a Kapton tape. This with the purpose of prevent discharge from both sides of the reactor [94]. In general, the surface configuration have shown to be more efficient when comparing the efficiencies of volume versus surface nanosecond pulsed DBD for the case of ethanol destruction [95]. Recently, new designs have emerged, such as the hybrid volume DBD (VDBD)- surface DBD (SDBD) reactor [96]. This hybrid reactor has been reported to show a clearly superior performance compared to the only VDBD and only SDBD when employed for Ozone generation. One should recall that the VDBD is the common configuration employed in most ozone generators in industry for water treatment.

4.5. Arc discharge reactors

When the discharge gap between the two electrodes is too small or high current is flowing in a strong electric field a gliding arc discharge is formed. The electrical discharge moves along the dimensions of the electrode, hence the name gliding arc discharge. The current can vary from a few amperes to a few hundred of amperes. Usually, two slant or semi-drum like electrodes are separated by a small distance at the base. The separation distance increases along the height of the electrodes. Sometimes the electrodes have a tapered or point end and are kept very near to each other with a very high potential. At the narrowest separation the arc is thermal, as the arc glides to the top it attains non-equilibrium characteristics. A typical schematic of gliding arc discharge is shown in figure 8. The gas is injected where the separation distance is in minimum. There are 3 phases in a gliding arc discharge which are formed from shorted to longest separation discharge: (1) gas breakdown phase, high voltage breaks the working gas to form a discharge, (2) equilibrium heating phase, the discharge achieves a higher bulk temperature due to its thermal plasma characteristics and (3) non-thermal plasma stage, formed as the arc moves upwards at longer separation distances [97]. The current profile with respect of dimensions can be tuned (in a limited range) with the flow rate of the gas. Arc plasmas are used for nanomaterials synthesis such as carbon nanotubes [98], boron nitride nanotubes [99], high pressure sodium lamps [100], decomposition of water pollutants [101], etc. Moreover, gliding arc discharge has been often pointed out as a plasma source with great scalability potential.

Zoom In Zoom Out Reset image size

4.6. Corona discharge

Non-thermal corona discharge occurs at atmospheric and high pressure in regions of sharply non-uniform electric fields. In this discharge, the field near one or both the electrodes must be stronger than in the rest of the gas. This occurs near sharp points, edges, or small-diameter wires, which tend to be low-power plasma sources limited by the onset of electrical breakdown of the gas [59]. This restriction can be avoided by using pulsating power supplies.

In corona discharges the intensity of the current is up to a few microamperes. Some arcing can be seen if another conductor comes in the vicinity of the high voltage electrode. To prevent the voltage leak, round tips are preferred as compared to conical tips. The electron temperature in a corona discharge exceeds 1eV with the gas at room temperature. The corona discharges are widely applied in polymer treatment. For example, synthetic fabrics to make clothing are treated before dying with corona-like discharges to provide enhanced adhesion. In general, the corona discharge is considered a major example of an atmospheric pressure non-thermal plasma source. Other important applications include precipitators [102], photocopying [103], water treatment [104], air ionizers [105] and pre-ionizers in lasers [106]. It is usually not preferred for producing chemicals in considerable quantities. Generally, Tesla coils are operated as corona discharges. A schematic is depicted in figure 9.

Zoom In Zoom Out Reset image size

Plasma is a very rich chemistry environment where different species are comprised in this ionized gas phase. From reaction kinetics perspective, important species that could contribute to chemical reactions present in a non-thermal plasma are ions, neutrals, vibrationally excited species, rotationally excited species, electronically excited species, surface adsorbed species and electrons. Hence, plasma catalysis becomes an extremely complex process. To provide an overview, the various interactions that take place in between the gas-phase species, wall and surface adsorbed species are summarized in table 3. The main purpose is to inform the reader about the complexity of a plasma catalytic process. The interactions in this table become more complex as the number of molecules and/or their complexity increases.

It is necessary to briefly introduce to the reader the different mechanisms of ionization which in general, can be divided in 5 groups:

• 1.  
  Direct ionization by electron impact is the ionization of neutral and non-excited atoms, radicals and molecules by an electron with the energy necessary to lead to ionization in one collision. This process is important in cold plasmas where the electric fields and electron temperatures are high, but the excitation of neutrals is moderate.
• 2.  
  Stepwise ionization by electron impact is the ionization of preliminary excited neutral species. This process is important in thermal discharges.
• 3.  
  Ionization by collision of heavy particles happens during ion-molecule or ion-atom collisions, as well as in collisions of electronically or vibrational excited species, and in this case the total energy of the collision partners exceeds the ionization potential.
• 4.  
  Photon-ionization happens during collisions of neutrals with photons resulting in the formation of an electron-ion pair. This process is important in thermal plasmas and in some cases in non-thermal plasmas when dealing with propagation mechanisms.
• 5.  
  Surface ionization/electron emission happens by electron, ion, and photon collisions with different surfaces or simply by surface heating.

It should be noted that the electron impact dissociation, can be considered as a pivotal reaction in plasma chemistry. This reaction is a function of the cross-section of the targeted atoms, hence, equation (1) is typically used to evaluate the rate of dissociation and cross-section to determine if an electron impact will result in dissociation or ionization. If the electron energy is more than the bond strength but less than the ionization potential, only then dissociation will occur [59].

$$\begin{eqnarray}&&{w}_{atom}={k}_{i}\left({T}_{e}\right)* {n}_{e}* {n}_{0}\end{eqnarray} \tag{ 1 }$$

where, ${w}_{atom}$ = rate of dissociation, ${k}_{i}\left({T}_{e}\right)$ = rate constant as a function of electron temperature, ${T}_{e}$ = electron temperature, n_e  = electron density, n₀ = particle density.

It is important to mention that positive ions are major players in plasma catalysis. Typically, their exothermic reactions with neutrals have no activation energy which makes their contribution significant in many processes. In this respect it is necessary now to look at the different mechanisms of electron ion recombination in plasma.

• 1.  
  Dissociative electron-ion recombination is known as the fastest electron neutralization mechanism in molecular gases or when molecular ions are present. In this process the recombination energy goes into the dissociation of the intermediate molecule ion and on to the excitation of the dissociation products.

  $$\begin{eqnarray}&&e+A{B}^{+}\to {\left(AB\right)}^{* }\to A+{B}^{* }\end{eqnarray} \tag{ 2 }$$
• 2.  
  Three body electron-ion recombination is the type of electron-ion neutralization that happens in atomic gases when there are not molecular ions present. The excess of energy in this case goes into the kinetic energy of the free electron, which participates as the 'third-body partner'. It is important to mention that ions and neutrals (heavy particles) are not effective as third body partners since they are unable to accumulate electron recombination energy fast enough.

  $$\begin{eqnarray}&&e+e+{A}^{+}\to e+{A}^{* }\end{eqnarray} \tag{ 3 }$$
• 3.  
  Radiative electron-ion recombination is the process where the recombination energy can be converted into radiation. The cross section of this process is relatively low and can compete with the three-body recombination when the plasma density is not high.

$$\begin{eqnarray}&&e+{A}^{+}\to {A}^{* }\to A+\hslash \omega \end{eqnarray} \tag{ 4 }$$

Finally, let's look briefly at the surface chemistry which plays a primary role in plasma catalysis. At this point it is important to recall the three main steps for a catalytic reaction to occur: 1) adsorption of reactants on the catalytic surface, b) surface reaction and c) desorption of products. From these, the surface reaction in plasma catalysis is the phenomena that has most unknows up to date. However, we can at least revisit some of the three prototypical thermal surface mechanisms.

• 1.  
  Langmuir-Hinshelwood in this mechanism the reaction proceeds between 2 species adsorbed on the catalytic surface. The resulting product molecule desorbs from the surface. This is the most common mechanism in thermal catalysis. The rate r of the reaction between two species (A and B) on the surface S is given by:

  $$\begin{eqnarray}&&r=k{\theta }_{A}{\theta }_{B}\end{eqnarray} \tag{ 5 }$$

  Where k is the rate constant of the overall reaction and θ_A and θ_B represent the surface coverage by the molecules A and B. The surface coverages can be correlated to the partial pressures of the molecules by the Langmuir isotherm:

  $$\begin{eqnarray}&&r=k\displaystyle \frac{{K}_{A}{p}_{A}\cdot {K}_{B}{p}_{B}}{{\left(1+{K}_{A}{p}_{A}+{K}_{B}{p}_{B}\right)}^{2}}\end{eqnarray} \tag{ 6 }$$

  Where ${K}_{A}$ and ${K}_{B}$ are the equilibrium constants for surface adsorption of molecules A and B and ${p}_{A}$ and ${p}_{B}$ are the partial pressures.
• 2.  
  Eley-Rideal in this mechanism, pre-adsorbed surface species react with an incoming gas phase specie, the resulting product molecule desorbs from the surface. The rate of reaction in this case is given by:

  $$\begin{eqnarray}&&r=k{\theta }_{A}{p}_{B}\end{eqnarray} \tag{ 7 }$$

  Where ${p}_{B}$ is the partial pressure of species B. Only very few reactions follow this mechanism.
• 3.  
  Mars and Van Krevelen this mechanism is characterized by the incorporation of one or more lattice constituents of the catalyst in the product molecules. The catalyst is then replenished by adsorption and uptake of corresponding species from the gas or plasma phase. Then in this case, the catalyst is temporarily consumed.

After the previous section on plasma discharges important in chemical processing, it is now necessary to understand how the plasma and a catalytic material can interact. The synergy of plasma catalysis means, that the combined effect of plasma catalysis can be higher than sum of individual effects of plasma and catalysis. Plasma catalysis is characterized by synergistic effects and sometimes, the effect of plasma catalysis can be even lower than the individual effects of only plasma and only catalyst. More details about synergy of plasma catalysis can be found in the literature [24, 107–109]. Plasma-catalysis itself can be considered as a variation of heterogeneous catalysis, where a solid or the catalytic material is immersed in a fluid. The gas species interact with the catalytic material by adsorption and weakly bound to the surface (physisorption) or they can dissociate and strongly bound to the surface (chemisorption). These species can react with other surface species following the Langmuir-Hinshelwood (L-H) mechanism or alternatively they can react with species from the gas phase following the Eley-Rideal (E-R) mechanism. This itself leads to one of the main differences with thermal catalysis where the reactive species are formed by dissociative adsorption on the surface of the catalyst, while in plasma catalysis the initial reactive species can be formed by: (1) dissociation within the plasma gas phase and (2) subsequent reaction of plasma gas generated species. Hence, this leads to the inference that a different start point will lead to a different reaction pathway. However, this process is more complex that just understanding the new environment or in this case plasma itself. It is the existing synergism between the catalytic surface and its surroundings i.e., plasma, that escalates the complexity of the plasma-catalysis process. Moreover, the lack of knowledge about the surface phenomena when a catalytic material is exposed to plasma, the missing understanding about the leading species in this new scenario and the proper selection of materials to exploit this new reaction pathways where plasma species play a prominent role limits this technology. This complexity can be elucidated by looking at the different interactions between a catalytic material and plasma. Specifically, the effects of plasma on the catalyst, the catalyst on plasma and the synergistic effects are summarized in figure 10.

Zoom In Zoom Out Reset image size

As it can be observed in the figure above the effects of plasma on the catalytic material include changes in the physicochemical properties of the catalyst such as high surface area [110], higher adsorption probability [111], change in the oxidation state [112], coke formation reduction [113] and change in the work function due to charge accumulation at the catalyst surface [114]. Other important effects for kinetic purposes of the plasma on the catalyst are the formation of hot spots, which can modify the local plasma chemistry. This can be exemplified by the case when metal nanoparticles on a support are packed in a plasma reactor. In this case, it has been observed that the heat transfer rate between the plasma and the small metal nanoparticles occurs faster than the heat transfer rate between the plasma and the support. Hence, there is an important degree of non-homogeneity with respect to time and position for the properties associated to the plasma-catalysis process such as: concentration of species (among them electrons), electron energies, electric fields and surface temperature. Finally, the possibility of a lower activation energy is another important effect the plasma can have on the catalyst. This is mainly due to the presence of short life-time species such as vibrationally excited species and radicals.

At the same time the catalyst and the packing can affect the plasma by possibly enhancing the local electric field in the plasma. In fact, it has been reported that there are enhanced electric fields in areas where ferroelectric packing beads or pellets touch each other (grain boundaries) and a higher electron density as well. As a result, this can lead to a change on the discharge type from streamers inside the plasma to streamers on the catalytic surface, hence a plasma intensification occurs at the contact points.

Moreover, the catalyst and catalytic packing can greatly influence the kind of species in the plasma gas phase, since some of these species can be adsorbed on their surface hence affecting the concentration of species in the plasma gas phase. At the same time this will have an effect in the residence time and simultaneously new species can be formed at the catalyst surface. This shows the importance of a good reactor design and the packing of the catalytic material.

An interesting challenge from the materials point of view is the formation of microdischarges in pores. While experimental results on pollutants removal from air showed that the discharge does not penetrate into nanoporous materials (<0.8 μm), this is doable in mesoporous materials (>15 μm) [31]. The possibility of microdischarges in pores has been studied in detail by modelling and simulations. It has been concluded that plasma streamers can penetrate into catalyst pores when the pore diameter is larger than the Debye length [115]. The Debye length radius is typically >3 μm in a microdischarge [116], however this value varies with the nature of the discharge. For example, filamentary discharges, e.g. in air, can have a high electron density in the streamers, yielding a small Debye length in the order of a 400 nm–1 μm. Hence, it is evident that there is an important dependence with the local electron temperature and density. Based on this, it has been pointed out the theoretical possibility that plasma streamers can penetrate into catalyst pores of several 100 nm diameter [49]. However, there is a long road ahead to prove this statement experimentally.

6.1. Plasma catalysis: an alternate approach to thermal catalysis?

Let's first focus our discussion onthermal catalysis, in order to contrast the two approaches. In traditional thermal catalysis, it is the use of heat that helps to activate the catalytic material for the reaction to proceed. Hence, typically harsh conditions are employed for such purpose such as in the Haber-Bosch process for ammonia synthesis where the temperature and pressure employed are 700 K and 100–200 atm [117]. Moreover, catalytic materials in thermal catalysis are considered non-active at mild conditions due the strong dependence of the surface chemistry with temperature. Furthermore, by changing the catalyst a completely different surface chemistry can occur and consequently a new reaction pathway can emerge. This simply means that we have very little control over selected or critical reaction steps by just varying the catalyst.

On the other hand, plasma catalysis can have the potential of being more selective than thermal catalysis since it offers the possibility of a rich gas phase chemistry. This rich gas chemical environment can open the way to new degrees of freedom non possible with just thermal activation where more homogeneity is achieved. As a result, this can lead to the selection and tailor of materials that benefit certain important reaction pathways such as in the case of photo catalysis, where electrons and holes play an important role in the surface reactions hence the catalytic materials are selected based on this surface interaction [118, 119]. Another example are plasmonic nanostructures that can channel the visible light energy in to specific states of adsorbed molecules [120–122]. That when populated, these metastable states can lead to higher rates of reaction compared to thermal catalysis. As a result, the reactions occur at milder conditions compared with only thermal activation. In the same way in plasma-catalysis by understanding the main reaction pathways and by determining the major species contributing to these reaction mechanisms one can select and tailor the catalytic materials accordingly.

In fact, the vast potential of plasma technology has already been proven by the commercialization of ozone synthesis, where this technology has essentially no competitors. Overtime, plasma-catalysis has become a benchmark technology for industrial-scale production of specialized materials as well. Silicon semiconductors grown by plasma enhanced chemical vapor deposition using silane and metallic catalysts are an example. Low temperature synthesis of graphene using hydrocarbon gas such as methane as precursor to obtain high quality uniform films on solid and molten metal catalysts is another example [123–126]. In recent years, plasma-catalytic air purifiers have been developed for commercial applications. Besides catalysis and fusion, this technology has several important applications. Plasma actuators have been developed for high precision flow control in reactors and aerodynamic foils alike. The synthesis of metallic nanoparticles is carried out by plasma troches as they can reach high temperature without external heating. Plasma sterilization has already been approved and accepted as a standard by the Center for Disease Control (CDC) in late 2000s. Though plasma televisions might be this technology major application. Moreover, plasmas are employed industrially as well as in research practices to clean and etch surfaces. In the past decade, plasma also has found application in treatment of human teeth [93, 127], skin [128–130] and even as a possible effective method for cancer treatment [131–135].

When employing plasma for catalysis purposes, the plasma-catalyst synergism might possibly lead to new reaction mechanisms and switch of the rate limiting steps compare to thermal catalysis. These promising features of plasma-catalyst interaction fit the criteria to inceptalternative eco-friendly routes for decentralizing several processes of great societal interest. However, to reach this goal it is necessary to understand the plasma itself from a fundamental and practical point of view and moreover, design plasma catalytic reactors that could have the capability for such processing. Also, we need to understand the synergy, we must be able to decouple the effects of plasma, catalyst, plasma on catalyst and catalyst on plasma before pilot-scale application of these so called 'Plasma Catalytic Reactors'. Thus, reactor design is an extremely important task which must be performed with outermost finesse for enabling the ease of operation as well as a mean to decouple these effects at the same time. Furthermore, plasma catalysis is a technique that can reconcile the use of alternate technology along with the use of renewable sources by coupling plasma reactors with renewable electricity sources such as solar and wind. Hence, plasma catalysis might offer the best way to store renewable electricity from intermittent sources and convert it into useful chemicals through highly selective reactions. Finally, it is important to list the main advantages that plasma catalysis can offer such as: (1) ultra-fast one pot reactions [107], (2) reactors can be adapted to renewable electricity sources since they can easily be switch on/off [136], (3) clean processes since they do not require the use of solvents, hence almost none or not waste is generated [64, 137, 138]. Figure 11 summarizes these advantages.

Zoom In Zoom Out Reset image size

Plasma diagnostics is a tool that help us to find some useful parameters to define and characterize the plasma. There are several important techniques employed, but the most extensively used world-wide can broadly classified into: (a) non-invasive testing and (b) probing techniques. Recently with progress in computing technology a third approach has been developed i.e. (c) simulations or modelling approach.

Non-invasive testing includes the use of instruments which are not in physical contact with the plasma. These techniques can be divided into two sub-categories viz. (1) in-situ and (2) operando methods. Usually, spectroscopy techniques are used for diagnostics. Emission spectra of plasma is collected through a system that contains a sensor. The most common technique for plasma spectroscopy is emission spectroscopy i.e. to capture the photons emitted by the plasma and then comparing with calibration values. The emission spectrum gives information about the species present in plasma along with their unique excited state by identifying the emission lines present in the emission spectrum.

Probing techniques include the insertion of tip like materials which are in contact with the gas-phase plasma. The most widely used probe for this purpose is called Langmuir probe. A Langmuir probe measures the current and voltage and can help in getting precise measurements for electrical properties of the plasma. The obtained data can then be used to predict the electron temperature and density. To closely monitor the electrical characteristics of plasma, Langmuir probes and high voltage probes paired to an oscilloscope are typically used. These probes can help in recording the changes in current and voltage with the progress of the reaction. With data analysis it is possible to calculate the changes in electric field, power consumption, plasma current, etc. Using the data from a Langmuir probe makes possible to calculate electron density and temperature, however one main disadvantage is that probe is immersed in the plasma which changes the plasma characteristics significantly in some cases. When using high voltage probes, it can become challenging to separate the imaginary current from the real leading to two types of electrical characteristics, i.e. one due to resistance of the reactor and another due to reactance. Some of the most outstanding techniques to obtain real-time operando plasma properties are molecular spectroscopy, i.e. emission spectroscopy, adsorption spectroscopy, laser induced breakdown spectroscopy (LIBS) and Raman spectroscopy. The major difference between these techniques is either the light source or the wavelength range. For example, Raman spectroscopy, LIBS and adsorption spectroscopy have an external light or laser source which bombards the plasma and a detector on the other side that collects the light to calculate the adsorption. Whereas for emission spectroscopy the photons emitted by the plasma are collected by the detector. These non-invasive light sensitive-spectroscopy techniques are also known as optical spectroscopies. Operando and in-situ characterization using the aforementioned techniques would help in transitioning from non-ideal to realistic values. The main advantages and disadvantages of these techniques are summarized in table 4. Simulations or modelling is another approach to diagnose plasma. In the past years several commercial software and individual codes have been developed to adapt curve fitting of different peaks and then run regression on peak ratios to find various parameters. Although, it should be noted that curve fitting applies only when an experimental OES spectrum is available for comparison. Most of these codes rely on peak broadening, line ratios or solving the Boltzmann equation (considering the system follows a Boltzmann-Maxwellian Electron Energy Distribution Function). This approach is usually used when either everything is known about the process or very little information is available.

For any gas-phase reaction, there are several current challenges faced in pioneering plasma-catalysis. One of them is the trade-off between conversion and energy yield. If we focus on a reaction with simple reactant molecules involved such as the plasma catalytic synthesis of ammonia this point can be clearly observed. In vacuum plasmas for example, there have been reported conversion values as high as 80% [20], however, energy yields are lower than 1 g-NH₃/kWh [4]. In the case of atmospheric-pressure plasmas the opposite is observed, energy yields are reported as high as 35.5 g-NH₃/kWh [141] with maximum conversion being less than 4% [46]. This major gap needs to be addressed by both process intensification as well as from materials point of view. Figure 12 [15] represents the highest values obtained for ammonia yields (%) versus energy yields (g-NH₃/kWh) by various groups using plasma-catalysis. It is evident from this graph that in radiofrequency low pressure plasma the energy yields are low while the percentage yields are high while the opposite is true for atmospheric pressure pulsed plasmas.

Zoom In Zoom Out Reset image size

Regarding the use of different catalysts so far, the kind of catalysts that have been explored in plasma catalysis are extremely limited, confined mainly to transition metals. Hence, another important current challenge to overcome is the tailor of catalysts for plasma environments and furthermore to understand their synergism with plasma. Currently, in order to further advance this plasma processing research field it is very important to understand the effects of plasma on catalyst, catalyst on plasma and plasma-catalyst synergy. Catalyst loading is another defy in plasma catalytic reactors. As of now either pellets or solid metals have been loaded. While this can be considered a trivial task in thermal catalysis, plasma catalysis requires this to be performed with extreme caution and understanding of the impact of the catalyst packing on the reaction itself. More efficient ways to incorporate molten catalysts, powders and thin films are needed, a task that can be possibly accomplished through a careful reactor design. Another drawback is limited in-situ and operando characterization of plasma and catalysts in plasma. As of now, emission spectroscopy, temperature contours and FTIR are the most common techniques. Plasma-catalyst synergy can only be understood if we understand the phenomena occurring at the surface, such as adsorption, diffusion, dissolution and recombination of species occurring simultaneously on the catalysts. Moreover, computational models coupled with experimentation have shown to be successful to unveil the possible plasma catalysis pathways. Hence this is a practice that can be further explored. Most importantly, the criteria for an ideal catalyst must be set which will depend on the fundamental knowledge gained through the development of this technique. While different discharges, unique reactor designs, and operation conditions are currently employed it is important to converge in generalities that can benefit the entire plasma community.

8.1. Towards the Ideal catalyst

8.2. Scaling up plasma catalytic processes

This tutorial is focused on summarizing the current basic knowledge on plasma catalysis aiming to offer a first brief insight to scientists wanting to explore this exciting field. The reader must know that a tremendous effort has been done up to date by the plasma community to further advance this field. However, there is still a long way to go in order to make possible the optimization and scalability of several plasma catalysis process. To achieve this, efforts should be done towards in-situ characterization of the surface of the catalyst during plasma exposure, effective coupling of experiment-simulation efforts and plasma reactors design. This latest task specially requires the utmost level of experimental skill and creativity to achieve enhanced catalytic performance and energy requirements. Hence, it is clear the need of multidisciplinary knowledge. The pathway ahead challenges every single aspect of chemical reaction engineering from the simple task of catalyst packing to the catalyst selection and design to determine reaction mechanisms and establish rate limiting steps for the new reaction pathways plasma offers. The following issues require special attention from the scientific community in the upcoming years: (This has been summarized in figure 13).

• 1.  
  Innovative design of highly efficient plasma reactors to allow the packing of different type of catalysts and pairing with renewable electricity sources. An effective reactor design can lead to a better contact of the plasma gas phase with the catalyst and to optimized plasma properties. Among them: space-time uniformity, gas temperature, and electron distribution function (EEDF).
• 2.  
  Selection and design of new generation catalysts for plasma environments. Currently, the materials employed as catalysts in plasma catalysis are similar to the ones employed in thermal catalysis where different reaction pathways occur. Hence it is needed to explore different class of materials to fully exploit the plasma-catalyst synergism. Moreover, it would be important to design/select low cost catalysts with high recyclability that can offer a real option for possible scalability.
• 3.  
  Understand the plasma chemistry at the interface of the catalyst. Despite the observed synergism between plasma and several materials employed as catalysts the knowledge of the plasma surface chemistry is very limited. This scarcity of data can be overcome with in-situ characterization techniques that can offer a more realistic picture of the phenomena happening in the reactor chamber. Moreover, it will be necessary an effective interaction between modelers and experimentalist to establish an efficient simulation-experimentation loop to gain a fundamental understanding to reach the full potential of plasma catalysis.
• 4.  
  Transfer of lab scale knowledge to industrial scale application. The possibility of transfer the knowledge we have gained so far in the lab to industrial scale for plasma reactors is a gigantic and challenging task that will need first to have solid fundamental knowledge of the reaction mechanisms. This challenge then will require the active collaboration from many disciplines ranging from materials to reactor design experts. While at the moment this can be seen as a long-term goal, the main driving force to pursue this task is the possibility of paring plasma reactors with renewable energy sources. Possibly making sustainable the synthesis of important chemicals.

Zoom In Zoom Out Reset image size

Maria L Carreon wants to thank NSF-CBET award No.1947303.