The Joshi effect

 The Joshi effect & Optogalvanic spectroscopy- historical perspective and applications 

Surya N. Thakur

Department of Physics, Banaras Hindu University, Varanasi-221005, India

1. Introduction

The Raman effect is widely celebrated as one of the most significant contributions by an Indian scientist, but the equally intriguing Joshi effect remains largely overlooked by many spectroscopists. While the Raman effect explains light scattering by molecules, the Joshi effect demonstrates how light causes a subtle change in discharge current within a low-pressure tube containing atoms and molecules. Discovered by Professor S.S. Joshi at Banaras Hindu University, the Joshi effect occurs when white light suppresses the glow in a discharge tube—termed the negative Joshi effect [1]. Conversely, the positive Joshi effect is observed as an increase in discharge current when high voltage is applied to electrode plates coated with compounds like mercuric chloride, sodium hydroxide, or potassium chloride.

Recent advancements in experimental techniques have renewed interest in studying atomic and molecular processes in discharge media, which are crucial for fields like astrophysics, plasma physics, and laser physics. For instance, in a CO2 gas laser system, various phenomena arise as  the laser beam interacts with gaseous mixtures. The simultaneous detection of photopyroelectric, photoacoustic and antenna effects have been reported in an intra-cavity of a conventional CO2 laser [2].  The antenna effect is an induction effect produced by the magnetic field generated as a result of the change in the discharge current during the lasing process. It can be detected by winding a coil around the laser tube at a point between laser electrodes. 

A related phenomenon is the optogalvanic effect (OGE), where light irradiation of a gas discharge alters its conductivity. The term "optogalvanic" merges "opto" (light) and "galvanic" (electric current). One extensively studied example is photoionization [3], where light induces electrical changes in the gas. Interestingly, the OGE also manifests as color shifts in the CO2 laser plasma tube [4].  Initially, the term "optogalvanic effect" emerged with the development of tunable lasers. Since then, optogalvanic spectroscopy has become an important non-destructive technique in laser spectroscopy. By directing laser light onto a gaseous medium, it identifies specific energy transitions in atomic species present in the discharge [5]. This method relies on resonance ionization, where atoms or molecules absorb photons at specific wavelengths, promoting electrons to higher energy states and causing measurable changes in electrical conductivity.

OGE was first discovered by Penning when one discharge tube was irradiated by another. In 1928, Penning and co-workers [6-10] illuminated a 20 Torr neon discharge with the emission from a similar discharge and observed an increase of some 10% in the voltage across the discharge tube. To demonstrate the effect graphically, the discharge tube is connected in a circuit containing a capacitor and a loudspeaker as illustrated in Fig. 1a. The pitch of the sound changes at the instant the discharge is illuminated by external light.

                                             

Fig. 1a Schematic diagram of the Penning experiment on the change of sound pitch of a loudspeaker upon illumination of a discharge: 1-discharge, 2-high-voltage source, 3-capacitor, 4-loudspeaker.  [Adapted from  Ref. # 84]  

  There has been a basic difference in the experimental observations of the Penning effect and the Joshi effect. In the Penning effect, the irradiation was from times prior to plasma formation in the discharge, and the increase in the threshold potential was explained as due to the destruction of the metastable states of neon in a Ar-Ne  gas mixture by the external neon light [6- 10]. The impact of external light on the Ar-Ne mixture in the discharge tube affects secondary ionization processes due to collisions of particles in the discharge. In contrast to the Penning effect, the discharge tube was irradiated after the formation of discharge plasma in the case of the Joshi effect [1], and not from times prior to discharge formation. Joshi effect was never observed in discharge tubes with metal electrodes, but observed only with the electrodeless discharge systems. The Joshi effect has been described as the positive or negative changes in discharge current in an electrode-less, high-voltage ac discharge-tube when irradiated by an external radiation. 

Joshi carried out his gas-phase reaction studies in a Siemens ozonizer, which consisted of a narrow annular gap between two coaxial glass tubes (see Fig. 1b). The novel feature of this discharge apparatus was that the electrodes were positioned outside the discharge chamber and were not in contact with the plasma. Joshi and Narasimhan [11] observed that the r.m.s. value of the stabilized discharge current, in an ozonizer filled with chlorine, diminishes almost instantaneously following the irradiation of the ozonizer by light from an incandescent lamp. This effect of external light is found to be almost instantaneous and reversible, and on stopping the irradiation the ‘dark’ value of the current restores in the ozonizer.  

It is important to point out that a few of the leading spectroscopists of the 20th century did recognize the importance of the Joshi effect in the context of Optogalvanic spectroscopy. Professor Herb Broida of the University of California at Santa Barbara was the first to point out "the close similarity" of optogalvanic spectroscopy with the Joshi effect. This was conveyed to Prof H.J. Arnikar by Dr. Kermit C. Smyth of the Institute for Applied Technology, NBS (now NIST), Washington D.C., during his visit in 1981 [12]. The importance of the Joshi effect in the context of Optogalvanic spectroscopy has also been recognized by Prof. G.W. Series of Oxford University [13] who states, “The first studies of the changes of the electrical characteristics of a discharge under irradiation by light were made by Penning in 1928, whose interest was in the changes brought about in the population of metastable atoms. It would appear that Joshi and his collaborators discovered the ‘light-effect’ independently, and not in connection with the role of metastable atoms. Apart from a passing reference to the effect by Joshi [1], the first specific account is by Joshi and Narasimhan [11]. A further account, with a suggested interpretation, is given by Joshi and Deo [14]”. 

   In the following sections we will discuss the experimental results and theoretical explanations of the Joshi effect followed by an account of several recent applications of the novel technique of optogalvanic spectroscopy.

2. Experimental studies on the Joshi effect

The experimental investigations that led to the Joshi effect, were the outcome of young Joshi’s fascination in a new area of silent electric discharge. This was also referred to as the  electrodeless discharge for effecting chemical reactions in the gas phase with high energies of activation. He used the silent electric discharge in Professor F.G. Donnan’s laboratory in the University College of London, to study the decomposition of nitrous oxide over a wide range of conditions [15]. In 1928, Joshi needed high voltage ac for his work in the Banaras Hindu University which was served by an outmoded 220 V, 200 kW dc supply in those days. He procured, from auction by Dehradun X-ray Institute, a host  of high voltage equipment including  many Westinghouse and Siemens 45 cm gap induction coils, Koch and Sterzel 40 kV transformers, 25 kV electrostatic voltmeters, large capacitors etc.  His work also needed plenty of glass-ware, tubes and capillaries, numerous vacuum stopcocks and gauges. Joshi practised the art of glass-blowing every day almost as a religion to train his co-workers in the same both at the table and in situ.

2.1 The silent electric discharge setup

Fig. 1b shows the general arrangement of the apparatus and the circuit employed for the study of reactions involving nitrous oxide, nitrogen and  phosphorus. A Siemens type ozoniser made with soft glass tubes was surrounded by a tube filled with salt solution to serve as a low tension electrode. The whole system was immersed in a jacket, in which a stream of water was kept flowing  at a constant rate in order to minimize the heat effects under the discharge. The ozoniser was connected through tap-1 to a mercury manometer and through tap-2 to a Toepler pump (see Fig. 1b).  After the entire assembly of the apparatus was tested for vacuum for at least 48 hours, taps 3 and 1 were closed. The inner electrode of the ozoniser was then coated uniformly on the outside with a concentrated solution of phosphorus in carbon di-sulphide, and fitted in the ozoniser system. The solvent was removed first with a hivac pump from tap-3 and finally from the mercury Toepler pump. At the commencement of every experiment, nitrous oxide from R1 and R2 was frozen in traps cooled by liquid air in D3 and D4; the middle fraction of the vaporised gas was led over a train of tubes charged with P2O5 and KOH, and introduced into the annular space of the ozoniser through tap-3 at the desired pressure [16].

Fig. 1b. Experimental setup to study the decomposition of nitrous oxide over a wide range of conditions. [Adapted from Ref.# 16]

The ac power obtained by means of a 1 kW rotary converter from 220 V, dc mains was fed to the primary of a 3 kVA transformer, with one of its secondaries earthed,  the other was connected serially through a 10 k stabilising resistance, while dipped in the salt solution to form the inner electrode of the ozoniser. The outer electrode was earthed through an ac micro-ammeter (A)  shown at the top of Fig, 1b. The applied ac  potential in kilo-volts (r.m.s.) called V, was calculated from a knowledge of the stepping up ratio of the transformer and of the primary potential V1 which was regulated by means of a variable resistance.  Joshi observed a discontinuous variation in the current at a given V during the decomposition of pure nitrous oxide under silent electric discharge [1].  In addition to providing data of analytical and kinetic interest, the work on reactions under silent electric discharge led to the following significant information:

• There is a characteristic threshold voltage Vb for the onset of each discharge-induced reaction. For a meaningful understanding of the reaction, the measured  potential across the electrodes had to be transformed into the potential effective along the mean free path of the reactant gas molecules.  

• Some systems under discharge displayed a periodic rise and fall in gas pressure accompanied by inverse rhythmic changes in current.  Thus, the direction of the following reaction was found to reverse once about every 20 minutes under certain conditions:                   N2O  +  H2    N2  + H2O    

• For potentials in excess of Vb, the emission of a faint glow as a broad band in the near UV-blue region, played no significant role beside the field-induced excitation-ionization collisions.  

• In the studies on active nitrogen for pressures around 40 Torr, the after-glow persisted indefinitely, but once quenched, its characteristic chemical reactivity also ceased.  Thus, there was no ‘glow-less’ form of active nitrogen  contrary to the then existing theories of active nitrogen.  

• The detection of an appreciable decrease in the discharge current was to establish itself as the Joshi effect.

2.2  Discovery of the Joshi effect

A puzzling observation was made during the study of the reaction  H2 + Cl2  2HCl, under silent electric discharge. As the reaction proceeds, chlorine of high electron affinity is replaced by hydrogen chloride of lower electron affinity. One would therefore expect an increase in the discharge current, on the contrary a decrease in the current was observed.  At the same time a faint glow was seen in the discharge space. The action of light, if any, would be to enhance the conductivity. To settle the issue of this strange behaviour, a discharge was set up in pure chlorine gas. After ageing for a short time till a steady V-i (current-voltage) characteristic was established, the gas under discharge was exposed to strong light from outside, and the current started falling rapidly to almost zero. On shutting off the external light, the current rose again to its former value. The  current suppression -i = (iL- iD) was practically instantaneous. This meant that the current in chlorine in-dark ( iD) at a constant potential was suppressed to a lower value (iL) under light and it formed one of the first reports as ‘A new light effect’ [11,14,17].

Fig. 2 The experimental setup and a variety of discharge cells used in the discovery of the Joshi effect.

A typical experimental arrangement used by earlier workers to observe the Joshi effect is shown in Fig. 2. The  discharge current could be monitored by the counter S or by the galvanometer G and its temporal profile could be seen on the cathode-ray oscilloscope. The inner electrode of the discharge tube was connected to the high voltage of the stabilized power supply and the outer electrode was grounded. The discharge current was measured by the counter S along the path shown by α, whereas its measurement by the galvanometer G was done along  β, and the measurements by the oscilloscope were done along γ. The counter recorded the amplified current, both in the absence and in the presence of external light on the discharge tube, after it was filtered by a capacitor. For current measurement by the galvanometer, the electrodes of the discharge tube were connected to a low frequency transformer and the current was rectified by a Sylvania crystal IN34. The temporal profile of the current signal was seen on the oscilloscope by using an appropriate carbon resistor to adjust its shape and the phase.   

The -i ‘light effect’ in chlorine was studied systematically over a wide range of conditions of applied potential, gas pressure, as well as intensity and wavelength of the irradiating light. The study was extended to several other gases and vapors and it soon transpired that the suppression of the discharge current -i, by the impact of external light, was a general phenomenon producible in most gases and vapors over a wide range of conditions [18,19].    

                               

Fig. 3 Experimental arrangement for the observation of the positive as well as negative Joshi effect in discharge through iodine vapor [Adapted from Ref.#19]

In halogen vapors, the magnitude of the Joshi effect was observed to vary in the order  Cl2>Br2>I2 and in the case of iodine vapor very little effect could be observed [17].  A positive Joshi effect (+i) was observed for the first time in iodine vapor at low potentials and a negative Joshi effect (-i) at higher potentials [18].  The observation of the Joshi effect was found to be extremely difficult without a sensitive detector and an adequate intensity of the external irradiation. Using a much improved experimental arrangement shown in Fig.3 [19], with the galvanometer detector suitably shunted by germanium crystal-diode 1N34 (Sylvania), both positive and negative effects were observed. The Siemens type ozoniser, containing powdered iodine in the bulb, was connected to a Toepler pump for evacuation.  The container was sealed off after being exposed to vacuum for more than 48 hours. The iodine vapor was in contact with the solid phase during the course of the experiments. The ozoniser was excited under various applied fields of 50 Hz frequency and the discharge current was monitored when the ozoniser was in the dark (iD), and when irradiated externally from an incandescent bulb (iL). The dark current (iD) increased, with increase in the exciting potential, rising to a maximum, then gradually diminishing and then continuously rising as illustrated in Fig. 4. The positive Joshi effect is observed at lower excitation potentials and the negative effect appears at higher potentials, and has been confirmed by the work of Khastgir and Setty [20, 21].  

 

Fig. 4 Effect of exciting potential on the discharge current (iD) when the ozoniser is in the dark (solid curve with open circles), and when irradiated (iL) externally from an incandescent lamp (dotted curve with  * ). The variation of the Joshi effect is illustrated by the dotted curve with solid circles. [Adapted from Ref.#21]

2.3 Mechanism of the Joshi effect in low frequency electrodeless discharge in chlorine

In the reports of Joshi and co-workers [22-24], when a low frequency electrodeless  discharge is established in an ozoniser filled with a halogen at 1-600 Torr, the r.m.s.-value of the discharge current decreases on irradiation by an incandescent lamp. However, no satisfactory explanation for the mechanism of this phenomenon has been developed.

Harries and von Engel [25, 26] were able to demonstrate the reduction in current pulses from a separate type of experimental setup. The current flowing through a 50 Hz electrodeless  discharge  in a short cylindrical glass tube fitted with plane electrodes at the ends and filled with chlorine at 10 Torr, was investigated with an oscilloscope. The system exhibited between 1 and 50 distinct pulses  per half-cycle, and the number increased with voltage. They were able to show that each pulse requires a large number of electron avalanches, which develop between opposite small areas of the inner glass surfaces. 

     The glass discharge tubes used by Harries and von Engel were of 6 cm diameter and 1-3 cm in length. The Cl2 gas at a  pressure of  10 Torr in the discharge tube excited by 50 Hz ac power supply, was investigated with a double-beam cathode-ray oscilloscope. The output current was found to consist of many distinct pulses and their number increased on increasing the exciting voltage. When the discharge cell was irradiated by a 40 W incandescent lamp from a distance of 50 cm, a reduction in pulse heights was observed.  At higher voltages this reduction became less, except for the first pulse of a group in each half-cycle. It was confirmed, by use of appropriate filters, that light of wavelengths less than 480 nm caused the reduction, corresponding to the dissociation energy of Cl2.

There is little doubt that the gas molecules and atoms, inside the discharge cell, form adsorbed layers on glass walls. The atoms will adhere by Langmuir adsorption sharing an electron with an alkali atom in the glass. The molecules are held mainly by van der Waals forces; so there will be a sparsely distributed population of atoms on the glass, with molecules on adjacent sites, which would be topped with several layers of molecules. This phase is akin to a liquid, and when irradiated  by external light, it produces a high concentration of atoms. In the experiments of Harries and von Engel it was found that the pulse heights were reduced by the preceding pulse, provided the pulses occurred within 10-4 sec of each other. This reduction is suggested to be due to metastable atoms (8 eV above ground level) and resonance radiation handed on from atom to atom, dissociating molecules in the wall layers.  

The current pulses did not change when irradiating only the region around the central axis of the discharge tube. On irradiating one wall of the discharge tube, the pulses were reduced every other half cycle only, namely those pulses whose electron avalanches started from that wall. Calculations showed that any Cl atoms formed near the center of the discharge tube would not recombine before reaching the walls, and hence the gas at the walls is the controlling factor [24].   

At room temperatures, chlorine molecules form an adsorbed layer, several molecules thick on the inner glass walls. Irradiation by external light produces photo-dissociation of Cl2, the atoms being effective in capturing electrons in this layer where slow electrons and many-body collisions are likely. This reduction of the number of secondary electrons reduces the pulse height. At higher voltages the interval between two successive pulses becomes so small that one pulse reduces the height of the following one. This is probably due to metastable Cl atoms and resonance radiation diffusing to the wall where they cause dissociation of molecules in the adsorbed layers, and the chlorine atoms capturing electrons as before. This also explains why at high voltages the effect of radiation becomes ineffective.

The secondary electrons from the wall that start the latter avalanches of a pulse can become attached to the Cl atoms. The probability of attachment to Cl2 is negligible. In the wall layer, the conditions for attachment are favourable because the velocity of electrons is low, and attachment is only likely in a "many-body" collision. Then the affinity energy (3.5 eV) can either dissociate the molecule forming the third body, or be converted into kinetic energy. Without a "third body" the excess energy can only be emitted as a quantum, which is rare. The loss of secondary electrons reduces the effective multiplication and hence the pulse height.

3. Theories and models to explain the Joshi effect

It is to be noted that the Joshi effect was never observed with the discharge tubes containing metal electrodes, it was observed only with  the electrode-less discharge systems. It was described as the positive or negative change in the current, in a discharge tube exhibiting a high-voltage electrodeless alternating current corona discharge, and  irradiated by external light.  Joshi envisaged three stages in the development of the current suppression:  

• The formation under discharge of an activated absorption layer at the glass-gas interface, consisting of gas molecules, excited and ionized atomic species and electrons.

• For the release of electrons from the absorption layer on external irradiation,  the work function being low, there is an increase of the discharge current  (+ i)

• The capture of the photo-electrons by atomic species of high electron affinity, in the many-body collisions in the absorption layer, results in current suppression (- i)  

  

These postulates were found to qualitatively explain the observed variation of -i with applied electric potential as well as the intensity and wavelength of the external light.

Parshad [27] provided an explanation of the Joshi effect based  purely on theoretical considerations. According to this model there would be a decrease  of the refractive index and hence the dielectric constant of the gas in the discharge under external irradiation. This in turn, should result in a decrease of the capacitive current originating from the ozonizer glass walls which act as condensers. The assumption that the current in a silent-electric discharge is wholly capacitative with no ohmic component is not consistent with experimental data of oscilloscopic analysis of the current and with the observation of large  -i  at potentials much higher than the characteristic threshold potential  Vb  and accompanied by a glow distinctly visible in dark.

Saha and Ghosh [28] advanced the electron surface-charge theory for the Joshi effect, which was based on the sequence of three stages  in the electrodeless discharge. The primary step is deposition of electrons on the dielectric glass during discharge. The second step is an extraction of some of these electrons from the glass surface, and their release into the gas phase during irradiation from an external light source. The final step involves the capture of a fraction of these electrons by the atomic species in the gas phase to form negative ions. This model attempts to compute the approximate values of the relevant parameters.  The surface charge density of 1010 electrons per cm2 gives rise to a surface potential of 107 V, and the rates of photodetachment of the electrons during irradiation and their capture by the atoms to form negative ions can be calculated.  The rate of capture of electrons by atoms in the discharge was calculated to be 108 sec-1 and 106 sec-1  in chlorine and oxygen respectively.  A co-occurrence of positive and negative Joshi effects is envisaged in this model. The net effect is +i if the rate of photodetachment of electrons from the surface exceeds the rate of negative ion formation, and in the reverse case which is more general - i results. 

The theoretical explanation of the Joshi effect by one of the leading discharge physicists, A. von Engel [25, 26] is also based on the three postulates, viz. formation of a multimolecular layer on the glass electrode surface,  photoemission there-from on irradiation and the attachment of the electrons to atomic species of high electron affinity. von Engel’s theory is more precise in detail and the treatment is quantitative based on the oscilloscopic analysis of the current during the potential variation in each half-cycle. von Engel’s model is particularly applicable to his experimental studies of -i  in chlorine exposed to light of wavelength less than 480 nm which photodissociates chlorine molecules yielding Cl atoms of high electron affinity. However, this theory does not account for the large -i, observed in gases like H2, N2, He and Hg vapor which have no absorption in the visible and whose electron affinities are very low.

3.1 Studies of the Joshi effect using oscilloscope  

                                                               

Fig. 5 Visualization of an ozonizer as a cylindrical capacitor where r is the radial distance from the common axis of the cylindrical electrodes of the capacitor. [Adapted from Ref# 29]

In a review article on the Joshi effect, Wagh and Deshpande [29] have pointed out that a proper understanding of the plasma formation is a prerequisite for the explanation of the Joshi effect, that is, of the effect of irradiation on the discharge current. They report that an ozonizer may be regarded as a cylindrical capacitor and that the geometry of the capacitor affects the process of formation of the discharge plasma in it. Using the cylindrical shape of the plasma container, they explain many aspects of the Joshi effect including the role of optical excitation. Their starting point is the  1971 experimental work by Venugoplan [30]  where certain important features of the Joshi effect  may be stated in the following words:   

“When the voltage across a discharge tube is gradually increased, under certain conditions and in certain regions, feeble and intermittent but successive discharges that are visible to the naked eye can be seen. Under this condition, illumination from outside can stop such discharge instantaneously; as soon as the illumination is cut-off, the discharge begins to start again. Its response to putting-on and shutting-off  of the external illumination is instantaneous and repetitive. When there is no visible glow discharge, illumination from outside initiates the  discharge instantaneously; as soon as the illumination is cut-off, the discharge stops, thus revealing that external radiation is under certain conditions an indispensable factor.”    

Some of the images from oscilloscopic studies of high frequency electrical discharge by Kher and Kelkar [31,32] are shown in Fig. 6 to understand the characteristics  of current pulses in metal electrodes and in electrodeless discharge systems. When the peak value of the applied voltage, under square wave excitation of  the discharge tube, is less than that leading to plasma formation in the discharge, a single current pip passes through the tube as illustrated in Fig. 6a. The duration of this pip was found to be  about 5RC, R being the resistance in series with the tube. The relaxation pulses in the metal electrode discharge system under its square wave excitation are shown in Fig. 6b, and the current pulse in an electrode-less system are shown in Fig. 6c [31].     

      

Fig. 6  Oscilloscopic presentation of current pip and pulses under square wave voltage excitation [Adapted from Ref# 29]

 

           

Fig. 7  Oscilloscope traces of current pulses  under sine wave voltage excitation. The peak value of the voltage in each case is above the threshold value. [Adapted from Ref# 29]        

The current pulses from the metal electrode discharge system, under its sine wave excitation are shown in Fig. 7a  [31].  When the peak voltage is slowly increased beyond the threshold Vm, under sine wave excitation, secondary pulses appear for an ozonizer discharge as shown in Fig. 7b. A part of Figs. 7a and 7b exhibit the stable waveform of current over a complete cycle, while Fig. 7c shows five discharge current pulses of the first half cycle as resolved by the expansion sweep of the cathode ray oscilloscope.

  The observation of the Joshi effect by Kher and Kelkar [31,32] has been reproduced in Fig. 8. The dark current waveform shown in Fig. 8a corresponds to a 2 kHz  exciting voltage of 360 V, and Fig. 8b  shows the waveform under illumination by white light from an incandescent lamp. It is to be noted that there is a decrease in the height of only the first pulse, but not of the secondary  pulses in Fig. 8b. This shows that only the major current pulse is light-sensitive.  Kher and Kelkar [31,32] have established that the ‘effective time’ for the Joshi effect is restricted to the duration of the first major pulse.    

           

Fig. 8  The Joshi effect in an ozonizer under its sine wave excitation at 2kc/s and 360 V [Adapted from Ref # 29]

The observational features of the Joshi effect in an electrode-less alternating current corona discharge can be summarized as follows [30, 33]:

• The applied voltage is at the threshold potential, Vb (the rms value of the applied ac voltage at which discharge is initiated in the gas) for about 30 minutes  to give the ozonizer an initial period of ageing, i.e., exposure to discharge.  The process of ageing is necessary to obtain reproducible results, and the current through the ozonizer has a steady state pattern.  Vb is practically independent of the frequency of the applied ac voltage.

• On increasing the applied voltage above Vb, the first large discharge current pulse occurs along with additional secondary pulses of smaller current rise which can be resolved by the sweep expansion device of the CRO. A triangular shape with rounded top is observed for the secondary pulses , but not for the first major pulse.

• The Joshi effect has been observed using Cl2, Br2, I2, NO2, O2, H2, air, organic vapors, N2O, H2O, HCl, SO2, N2, rare gases, as well as vapors of Hg, Se, Te, Zn, Na, K, and Cu2O.

• Increase of applied peak voltage and that of its frequency diminish the magnitude of the Joshi effect. It is small under dc excitation.

• With the increase of gas pressure, i first increases to a maximum value and then decreases.

• The effect of rise in temperature of the discharge system is to increase the positive Joshi effect and to inhibit the negative Joshi effect.

• The magnitude and the sign of the Joshi effect are affected by electrode coating or filming and also by continued discharge. This indicates that the electrode-gas interface is the seat of this phenomenon.

The Joshi effect can be quantified by the percentage change i/i  with respect to the peak value of the dark current pulse. Thus,  we have the quantification of the Joshi effect as:          

                                                                                             

                                                       i/i = (iD-iL)/iD         (1)                                    

where iD is the dark discharge current and iL is the discharge current under irradiation from an external light source. Then, i>0 is called the positive Joshi effect and i<0, the negative Joshi effect.

3.1.1  Plasma formation in an ozonizer discharge

An ozonizer may be assumed as a cylindrical capacitor, and the electric field between the electrodes of the ozonizer is given by :  

                                    E(r,t) = V(t) / r ln(rout / rin)      (2)

where  V(t) = V0 sin t is the instantaneous value of the applied ac voltage and

                                                rin   r   rout  (3)

where r is the radial distance from the common axis  of the cylindrical electrodes of the capacitor (see Fig. 3).

If Es is the dielectric strength of the gas contained between the electrodes, then the voltage Vb the breakdown threshold, at which discharge  begins in the gas is given by

Vb = Es rin ln(rout / rin)   (4)

The breakdown of the gas begins at r = rin at V= Vb, and with increase in the applied voltage beyond Vb  to its peak value V0, the breakdown of the homogeneous gas-dielectric extends to the radius rb given by       

rb = V0/[Es ln(rout / rin) =  rin (V0/Vb) (5)

A breakdown zone only around the inner tube of the ozonizer has a thickness given by

r(t) = r - rin =  rin [V(t) -Vb] /Vb    (6) 

The maximum value of the thickness of the breakdown zone for a chosen peak value V0 of the applied voltage is given by

rmax = rb - rin = rin [V0 -Vb] /Vb           (7 )

It is important to note that the breakdown zone is not a surface layer because rmax can be of the order of a centimetre for V0 = 1.5xVb.   

Let us assume that the elemental thickness of the breakdown zone corresponding to the breakdown voltage V0 = Vb is re. Since the dielectric breakdown zone is centered around the inner tube of the  ozonizer, negative charges are rapidly swept away by the inner tube when it is acting as a cathode under the applied ac voltage (see Fig. 5).  The charge on the cathode is  rapidly lowered by Q, and consequently the potential across the electrodes is lowered by    V= -Q/C.  

Since the positive charges move to the anode through the unbroken dielectric, the moving front of positive charges undergoes variable acceleration in the electric field E(r,t). Thus, the moving front of positive charges drifts at varying velocity to the anode while covering a distance  rout - (rin + re); and the current pulse exists for the corresponding time. The amount of positive charge  Q now gets deposited at the anode, lowering the potential further by V  instantly.

In view of the above discussion, the dielectric breakdown within the zone of thickness  near the inner tube is accompanied by voltage changes across electrodes by 2V with the associated sharp and large major current pulse.

If the angular frequency , of the applied voltage, is such that the next front of charges is created before the earlier one has reached the anode, a new disturbance of pressure will simultaneously travel through the dielectric. Thus, there are different kinds of disturbances creating pressure waves, travelling at the speed of sound vs within the dielectric (gas).  Ageing  of the ozonizer is required to stabilise  these processes and to obtain a stable waveform for the current in the ozonizer. The ageing time depends on the frequency , the length rout -rin, as well as on viscosity of the gas.   

Fig. 9  Standing waves of pressure in an ozonizer

The stabilised pattern of the processes, described in the previous paragraph, will correspond to standing waves of sound within the cylindrical geometry of the capacitor (the ozonizer) as shown in Fig. 9. Major creation of charges due to secondary ionization will then occur at the corresponding pressure maxima (the antinodes of the standing waves).

Let us consider a stabilized waveform after the ageing of the ozonizer has taken place. The inner tube would act alternately as an anode and a cathode during each cycle of the applied ac voltage. The major pulse of current is provided by the dielectric breakdown near the inner tube of the ozonizer. It corresponds to primary charges of current within the ozonizer, and the charges cause further ionization outside the dielectric breakdown-zone. The fundamental mode of standing waves will be the first to be excited in the (dielectric) gas and the first secondary pulse of current will correspond to the collisional ionization at its pressure maximum (antinode) by primary charges. This first secondary pulse will have a symmetric triangular form since the antinode is located halfway between the electrodes (see Fig. 9). With the slow increase of the peak voltage V0 beyond the breakdown threshold Vb, a subsequent batch of secondary pulses will occur from the corresponding higher modes of standing waves (Fig. 9).

The above theoretical features of current pulses agree with the observations of Kher and Kelkar [31, 32] as reproduced in Figs. 7a and 7b.  Kher and Kelkar [31] provide the following empirical expression for the number of secondary current pulses corresponding to the maximum excitation voltage V0 .  

N = (V0 - Ve) /(Vi - Ve) (10)

where Vi is the ignition voltage at which the current pulse begins and Ve is the extinction voltage at which it ends. Eq. 10 has been substantiated by the theory developed here by Wagh and Deshpande [29].

It is important to note that the charge carriers of current in the ozonizer are positive when the inner tube is acting  as a cathode, while they are negative when the inner tube is acting as an anode. The characteristics of the discharge current pulses in the  ‘upper half’ cycle and those in the ‘lower half’ are due to different charge carriers. This difference in charge carriers is clearly recognizable from the differences in the observed pulses shown in Fig. 7. Thus in Fig. 7b for the upper half cycle of the current waveform, the rising portion  has electrons as dominant charge carriers and the falling portion has ions as dominant charge  carriers. However, for the lower half cycle electrons are the dominant charge carriers. Thus, the amplitudes of current pulses are larger during the lower half cycle than in the upper half cycle.  

3.1.2 The Joshi effect 

The process of plasma formation in an electric discharge is characterized by the following aspects [34-36]:

• A moving electron causes ionization in the gas with Townsend’s first ionization coefficient  ,  and the corresponding current is given by i  as

                         i = i0 exp(d)   (11)

where d is the size of the discharge gap and i0 is the initial value of current in the system.

• Secondary electrons are produced in many ways within a self-sustained discharge and all of them can be described by a single Townsend’s second ionization coefficient   . The secondary electrons produced per positive ion and the corresponding discharge current i, is given by

        i = i0exp(d) /[1- {exp(d)-1}]         (12)

The denominator of Eq. 12 becomes close to zero and very large current flows through the system at high voltage, which signals the dielectric breakdown.  Subsequently, the electric discharge maintains itself at the breakdown value of the voltage and at values higher than it.

A +i may now be interpreted to be the decrease in the breakdown voltage of the dielectric and a -i to an increase in the breakdown voltage. This can happen, in general due to increase or decrease of or or both. However, is not known to vary much, and the change in is the main determining factor for the change in the discharge current. The following mechanisms of removing electrons from the discharge space are expected to greatly affect the change of current through the ozonizer:

• Adsorption of molecules on electrode surface decreases    by increasing the effective work function of the surface,  which decreases the discharge current.

• Electron attachment to atomic species of high electron affinity also lowers the discharge current, because  mobility of charges decreases in  such an attachment.

• Electrons get removed from the discharge space in electron-ion  recombination processes which also lowers the discharge  current. In gaseous mixtures, electrons attach themselves to molecules of higher electron-ion recombination coefficient.

• In collisions between an ion and a  neutral molecule, charge transfer reaction can take place such that  A+ + B   A + B+. Appreciable number of electrons can be removed from the discharge space if the B+ ion  has much higher electron-ion recombination coefficient than the A+ ion. Thus, the charge-transfer process decreases the discharge current. 

An external illumination of the plasma container changes the conditions of the equilibrium under the above mentioned processes. As an example, the process of collisional ionization gets inhibited by an incident light quantum due to de-excitation of the atomic state in the following manner:

A* + h  A `       (13)

This process leads to a decrease in the discharge current under illumination. Under external illumination, new values of the equilibrium parameters of the dielectric (gas) get established. These new equilibrium values now correspond to increased or decreased values for the discharge current pulse. It has been found that only the first major current pulse is light sensitive as exhibited in Fig. 8. It is also important to remember that irradiation does not affect the process of secondary collisional ionization at the antinodes of the standing waves of pressure within the ozonizer. It is now clear that irradiation of external light affects only the discharge processes taking place within the zone of dielectric breakdown close to the inner tube of the ozonizer.

The reasons for only the first major current pulse showing the Joshi effect, and secondary pulses being insensitive to light, involve metastable states of atomic species. The Joshi effect is limited to the breakdown zone centered only around the inner tube of ozonizer that harbours the origin of the first major current pulse.

4. The Optogalvanic effect

In the introductory section of this chapter we have mentioned that the optogalvanic effect (OGE) was first reported by Penning [6-8]  and it is referred to as the Penning effect. It involves an increase in the threshold potential of argon-neon mixtures when irradiated by neon light. It has been explained as due to destruction of the neon metastables by neon light which inhibits secondary ionization by collisions. Neon has an excited metastable state Ne* (3P2 or 1s5), of energy 16.62 eV and of radiative lifetime to ground state of the order of seconds, and it plays an important role in collisional ionization of argon atoms as well as atoms of the cathode material.

Ne*(16.62 eV) + Ar     Ne (0 eV) + Ar+(15.8 eV) + e-       (14)

where 15.8 eV is the ionization potential of Ar atom

Similarly for the collisional ionization of atoms (M) of the cathode material

Ne* + M    Ne + M+ + e-                                              (15)

When the discharge tube, containing Ne-Ar mixture, is irradiated by light of photon energy h, from another Ne discharge tube, the high energy metastable species get depopulated due to the following processes:

  Ne*(16.62 eV) +  h Ne**(18.38…18.97 eV) Ne (0 eV) + energy (lost)     (16)  

where Ne** represents a group of closely spaced higher  excited states  (non-metastable) of Ne.

The atomic processes of Eq.16 inhibit the collisional ionization processes and hence the breakdown threshold (VL) for Ne-Ar mixture under irradiation by neon light rises above its value (VD) in the absence of external light (i.e.  in dark). Thus, V L-VD   is the Penning effect.   

A variety of studies have been carried out on the voltage changes occurring when an atomic species in a discharge is irradiated with photons whose frequency matches an electronic transition [37-40]. The change in the voltage between the electrodes of a gas discharge plasma when illuminated at wavelengths corresponding to allowed transitions in the gas atoms has been termed as the optogalvanic effect.  In a major development for understanding the optogalvanic effect, sophisticated measurements of laser-induced ionization changes were investigated by a dedicated group of scientists, mostly from the US National Bureau of Standards (now NIST) [41-47]. It was found that the millisecond pulse irradiation of the discharge in a hollow=cathode lamp with a cw tunable laser produces potential or current changes at wavelengths corresponding to electronic transitions in the atomic species under discharge. The V (or i) changes, recordable on an oscilloscope, accompanying and correlated to the spectral transitions, are termed the Optogalvanic effect (OGE)

4.1 Optogalvanic spectroscopy

The technique of optogalvanic spectroscopy monitors the changes in the discharge voltage  under irradiation from a wavelength tunable laser. In a sense, this is spectroscopy without a spectroscope, as it does away with the optical monitoring of the irradiating light as well as of the re-emitted light.

Optogalvanic spectroscopy is performed equally well by using a pulsed laser or a chopped cw laser beam for exciting the OG signals in the discharge. The experimental arrangement of Fig. 10 refers to a cw laser source where the laser beam can be modulated by a mechanical chopper at various frequencies up to 2 kHz. About 20% of the laser beam (10-20 mW) is directed axially into a commercial sodium hollow cathode lamp with neon gas at 5 Torr. The 2 kHz OG signals arising from optical resonances with electronic transitions in the discharge are separated from the  100 V dc anode potential in the neon lamp by using a capacitor in series with the current limiting resistance. The observed OG signals are relatively independent of the chopping frequency from 0 - 2 kHz and are recorded either with an oscilloscope or a lock-in amplifier. In the case of pulsed laser excitation, the OG signals are monitored with an oscilloscope or a boxcar with gated electronics.

                 

Fig. 10  Block diagram of the experimental arrangement for studies on the optogalvanic effect in hollow cathode discharge irradiated by a periodically chopped light beam from a tunable laser. [Adapted from Ref # 44]

4.2 Theory of optogalvanic effect

Erez et al. [48] have used a nitrogen pumped dye laser to excite atoms inside a homemade uranium hollow cathode lamp containing argon or neon.  The voltage transients arising from the absorptions of dye laser pulses are displayed directly  on the oscilloscope to reveal the temporal profiles of the OG signals.  Several excited states of atoms are populated in the hollow cathode discharge and optical absorptions from them give rise to their characteristic temporal profiles. The shape of OG signals in argon atoms as well as in neon atoms,  at different values of gas pressure and of the discharge current, are shown in Figs. 11 and 12. It is found that the OG signals originating from non-metastable Ar atoms exhibit only a negative voltage transient (see Fig. 11). On the other hand, the OG signals from metastable Ne atoms are composed of negative as well as positive voltage transients (see Fig. 12).

Erez et al. [48] have presented a phenomenological theory to explain the  magnitude, sign and the time evolution of experimentally recorded OG signals from the metastable and non-metastable levels of atoms.  They consider a dc discharge tube in series with a ballast resistor. The pressure of the buffer gas within the tube and the current are in the regime which corresponds to a positive internal tube resistance. A single electron emitted from the cathode is assumed to produce an avalanche of electrons with a multiplication factor  .  In the case of steady state discharge  =1. For <1 there is a decrease of the discharge current and for >1 there is an increase in the discharge current.  The change in the discharge current is followed by an opposite change (self regulation) in the voltage across the ballast resistor to restore the steady state. When a laser pulse is introduced into the discharge tube operating in a constant current mode,   it is assumed to produce a    quasi-steady state such that  1 and d 0. The OG signal is the small voltage deviation  V from the steady state, and it is given by the following expression:

                         V =  - ai ni    (17)

where ni are the various atomic and ionic populations and ni are their deviations from steady state.  = (/V)-1  and  ai = /ni .       

Fig. 11 OG signals from pulsed laser excitation of Ar atoms, at 588.86 nm, from the non metastable level at 105463 cm-1. (a) at different pressures. (b) at different discharge currents.[Adapted from Ref. # 48]

Suppose the OG signal is produced by optical absorption E1E2 (>E1) between a pair of atomic levels with population density n1 and n2  respectively.   According to Erez theory the temporal evolution of the OG signal is  given by the following expression:

V = -Q(n1-n2)[a2exp(-t/T2) - a1exp(-t/T1)]           (18)

                                                n1 > n2   and a2 > a1

where Q is a constant corresponding to the optical cross section and the intensity of the laser pulse,  T2 and T1 are the energy relaxation times for atomic states  E2 and E1 respectively.   

It is evident from Eq. 18 that  V(t=0) < 0 and the OG signal is always initially negative. For a pair of non metastable states T1 T2 and the OG signal decays exponentially in a time characteristic of the discharge plasma. For metastable states T1 >T2; the OG signal crosses zeo at a time t0 to become positive and then decays exponentially with a time constant T1, and   t0 is given by a2exp(-t0/T2) =  a1exp(-t0/T1). The positive OG signal has its maximum at t= tm  where tm  is given by  (a2/T2)exp(-tm/T2) =  (a1/T1)exp(-tm/T1). The temporal behaviour of the OG signal can be deduced from Eq.18 for V. The factor Q(n1-n2) is positive and does not affect the temporal profile of the OG signal which can be calculated from the factor [a2exp(-t/T2) - a1exp(-t/T1)].

  

Fig. 12 OG signals from pulsed laser excitation  of Ne atoms, at 588.19  nm, from the metastable level at 134043  cm-1. (a) at different pressures. (b) at different discharge currents.[Adapted from Ref. # 48]

The OG signals at 588.86 nm, shown in Fig.11, for Ar result due to excitation from a non metastable state and in view of the above discussion  its temporal profile is negative. It is  seen that the relaxation time is faster for higher currents. Increasing the gas pressure for a fixed current also indicates a faster but less pronounced relaxation time  at higher pressures.  The OG signals at 588.19 nm, displayed in Fig. 12 originate due to excitation from a metastable level in Ne, and its temporal profile has a negative and a positive component. At higher gas pressures the positive part of the OG signal is relatively smaller. The reduction in relaxation time with increased current is also evident.

5. Applications of the optogalvanic effect

The optogalvanic technique can be considered to be an alternative to the techniques of absorption or fluorescence spectroscopy. The experimentally observed laser-induced changes in discharge characteristics are usually small, and the OG effect can be considered directly proportional to the number of the photons absorbed. In a hollow cathode discharge, the OG effect makes it possible to perform spectroscopy on gas phase sample of refractory elements sputtered from the cathode.There is an wide range of applications of this simple experimental technique including trace concentration detection, isotopic analysis, laser-wavelength calibration and laser stabilization. In the following sections we summarise some of the important applications of the OG effect.

5.1 Calibration of wavelength of tunable lasers

Fig. 13 Spectral profile of the dye laser output in the wavelength region 615 - 675 nm (top) and OG spectral lines from a Fe-Ne hollow cathode lamp for calibrating the observed spectra produced by the wavelength tunable dye laser. [Adapted from Ref # 49]

Optogalvanic spectral lines provide one of the simplest methods of laser wavelength calibration in dye laser spectroscopy. This can be done in two steps; in the first step, a small fraction of the dye laser beam is incident on a hollow cathode lamp working in the constant current regime and the small changes in the discharge voltage is fed to one of the strip-chart recorder. The output of the laser power meter is fed to a second strp-chart recorder which simultaneously records the spectral profile of the dye laser as a function of the wavelength as illustrated in Fig. 13. The dye laser profile shown in Fig 13 has been recorded as the photoacoustic signal from a carbon black sample instead of a power meter. This spectral data may be used to normalize the spectrum of the sample in a later experiment where the laser-induced fluorescence or optogalvanic  signals are recorded simultaneously along with the signals from the hollow-cathode lamp.  

In one of the earliest experiments on optogalvanic spectroscopy, King et al. [44] studied the OG spectrum from a Na-Ne hollow cathode lamp operating at 10 mA while scanning the cw dye laser from 585 nm to 595 nm. The laser beam was modulated at 2 kHz by a mechanical chopper and the subsequent  OG signals were separated from the 100 V dc anode potential as illustrated in Fig. 10. The positive OG signals in Fig 14 correspond to optical transitions originating from the 2p53s (3P2,1,0) states, two of which (3P2,0) are metastable. Illumination of the electric discharge, which leads to a decrease in the collisional ionization rate due to the net depopulation of the high-lying metastable states, produces a positive OG signal.  Negative OG signals correspond to optical transitions which enhance the net collisional ionization rate in the discharge.

Fig. 14b shows an example of calibrating the linewidth of a tunable dye laser. Line widths of atomic transitions within hollow-cathode lamps are typically 0.004 nm [44]. The OG signal in Ne at 588.1895  was obtained from 18% of a 120 mW cw dye laser chopped at 2 kHz. Since the Ne linewidth  from the hollow cathode lamp is about 0.004 nm and the observed linewidth from Fig. 14b is 0.016 nm, the dye laser bandwidth is directly obtained as 0.016 nm.

 

Fig. 14 (a) Optogalvanic spectral lines of Ne in a Na-Ne hollow cathode discharge, except the two starred  lines in the lower trace. (b) Laser bandwidth calibration of the Ne line at  588.1895 nm using a single scan with 10 mW cw laser power. [Adapted from Ref # 44]

  Levesque et al. [50] have performed wavelength calibration of pulsed dye lasers using  uranium or neon atomic absorption lines detected by the OG effect in a hollow-cathode discharge.  They suggest that all of the parameters in the OG recording should be kept constant for a reproducible calibration. The highest wavelength calibration accuracy was obtained by setting the boxcar parameters to minimize line shifting. An accuracy of better than 0.005 nm in the wavelength calibration of a 0.02 nm FWHM dye laser was achieved.

     The reactive molecules and radicals, present in the atmosphere, can be easily produced in the electric discharge in laboratory conditions. Webster [51] has found a special technique based on  optogalvanic effect for wavelength calibration  of such reactive atmospheric species. In these investigations, the OG spectra of molecules observed in the discharge are compared with the atmospheric fluorescence spectra produced by the same laser beam.

5.2  Doppler-free optogalvanic spectroscopy

In the technique of intermodulated  optogalvanic spectroscopy (IMOGS), the output of a single mode dye laser is split into two beams of equal intensity and sent through a positive column of discharge in opposite directions. Lawler et al. [52] were one of the first to introduce the technique of IMOGS where the two counter-propagating laser beams are chopped at two different frequencies f1 and f2 while the OG signal is synchronously detected at a frequency f1+f2. A characteristic of this method is the appearance of a crossover signal when two spectral lines with a common level fall inside one Doppler width. In this case some atoms have a velocity component in the beam direction such that they are in resonance with photons from both of the beams because of an opposite Doppler shift  (see Fig 15).

             

Fig. 15  Intermodulated optogalvanic spectrum exhibiting a part of 23P-33D transition at 587.5 nm in 3He [Adapted from Ref # 53]

Hansch et al. [54] have developed the method of polarization-intermodulated-excitation (POLINEX) for optogalvanic spectroscopy where polarization of one or both of the beams is modulated rather than the intensity of the beams as in IMOGS technique. By modulating the polarization rather than the intensity, the detected Doppler-free signals arise from light-induced atomic orientation such that atoms which have suffered velocity-changing collisions will no longer contribute a signal if their orientation are destroyed. POLINEX technique takes advantage of the selection rules for the absorption of polarized light to yield information  on the angular momenta of the participating energy levels, as well as on the relaxation of light-induced alignment. An experimental arrangement for POLINEX is shown in Fig. 16 where only one of the beams is polarization  modulated, and the resulting OG spectrum of Ne is shown in Fig. 16a. The  intermodulated  fluorescence spectrum of Ne recorded by Hansch et al. [54] is also shown in Fig. 16b to illustrate the importance of the POLINEX technique in the reduction of the broad pedestal below the sharp Doppler-free peak present in the fluorescence spectrum.

Fig. 16 The experimental setup for POLINEX technique in OG spectroscopy along with (a) the Doppler-free POLINEX spectrum of Ne 1s5-2p2 transition. (b) The same spectrum recorded by the intermodulated fluorescence technique. [Adapted from Ref # 53]

An advantage of POLINEX technique, over the intermodulated method, is that any modulation of the excitation rate produces a nonlinear signal, and  good selectivity for the Doppler-free signal is maintained even when one polarization  modulation  is removed. The fact that the cross section for disorientation of atoms is usually of the same order as the velocity-changing collisions, also eliminates the broad pedestal present in intermodulated spectroscopy. A second advantage of POLINEX technique is that the crossover signal  is frequently of opposite sign, so that it can easily be distinguished in complicated spectra. This fact has been illuminated in Fig. 17 for the fine structure associated with the 23P-33D transition of 3He at 587.5 nm. It is to be noted that if an amplitude  modulation is present together with the polarization modulation, a spurious signal is produced.        

      

                                     

Fig. 17 Comparison of POLINEX optogalvanic spectroscopy with IMOGS. The upper part is a section of the hyperfine spectrum of the 23P-33D line of 3He and the lower part is the same spectrum recorded by IMGOS. [Adapted from Ref # 55]

                      

Fig. 18  Optogalvanic spectrum of He2 dimer. (A) Partial low resolution spectrum of the  b3g-f3u transition. (B)  High-resolution Doppler-limited spectrum of the R(4) and R (5) components. (C) Doppler-free intermodulated spectrum of the R(6) component. [Adapted from Ref # 55]

He2 is the largest known molecule of two atoms in the ground state with a bond length of 52 A, and this bond is 5000 times weaker than the covalent bond in H2 molecule. The three lowest triplet states of He2 are a3u, b3g and c3g, and the excimer molecules are more tightly bound than the van der Waals bonded helium dimer.  The spectrum of He2 excimer exhibits several bands due to a number of transitions between different electronic states

 Kawakita et al. [56] have studied the Doppler-free  intermodulated OG spectra of helium dimer molecule originating in  b3g  state at 573.2 nm (17446 cm-1) as shown in Fig, 18. The analysis of the triplet splittings of rotational lines provide values of spin-orbit and spin-spin coupling constants for the b 3 g and the f 3u states.

5.3 Optogalvanic photodetachment spectroscopy  

The optogalvanic technique has been successfully employed in diagnosing a discharge in which negative ions are present. The OG signal results from photodetachment of the electron from the negative ion, induced by the laser radiation when the photon energy is greater than the photodetachment energy. The photodetachment process does not change the number of free charged particles, but the discharge impedance is greatly reduced due to the very different mobilities of free electrons and negative ions. The OG signal in this case is easily distinguished from the signal due to molecular absorption by its spatial and temporal characteristics. Using a tunable laser, photodetachment OG spectroscopy can be employed to determine  the electron affinity of atoms  or radicals. Webster et al. [57,58] observed a sudden increase in the discharge current at the energy threshold of the photodetachment reaction in iodine when the laser was tuned from 420  to 380 nm as illustrated in Fig. 19. The threshold energy for photodetachment was determined to be 3.0591 0.0001 eV, with a simple and inexpensive discharge tube, and it was more accurate than the complex determination made with a crossed-beam method. Klein et al. [59] have used a frequency-doubled dye laser to determine the photodetachment threshold for CN radical in the 330 - 320 nm region, and the value of electron affinity  is found to be 3.821 0.004 eV. The photodetachment Optogalvanic technique has also been applied to the study of negative-ion kinetics in an rf discharge. Touzeau et al. [60] have monitored the density of O - in a pure O2 rf discharge.    

                      

Fig. 19 Photodetachment optogalvanic spectrum of iodine near the 405 nm threshold. The lower trace shows the potassium laser induced fluorescence (LIF) spectrum for wavelength calibration. [Adapted from Ref # 57]

5.4 Optogalvanic stabilization of laser frequency

The OG signal arises due to impedance change in a discharge cell with excitation by external light.  The standard technique of stabilizing the laser frequency is based on the impedance change in the excitation discharge as a function of the output power of the laser. The laser working in a single mode and in single-line regime is frequency modulated by tuning the length of the laser cavity using a piezoelectric transducer (PZT) as shown in Fig. 20. The OG signal is detected as an ac voltage across the ballast resistor of the dc power supply line by a phase sensitive lock-in detector. The output of the phase detector is amplified and sent to drive the PZT to correct the length of the laser cavity (see Fig. 20). A stabilization on the peak laser power of the order of 5x10-8 has been obtained by using this technique [61, 62].

Fig. 20 Schematic representation of high-accuracy optogalvanic stabilization of CO laser. [Adapted from Ref # 55]

      Moffatt and Smith [63] explained the continuous decrease in the OG signal and its ultimate disappearance, in low pressure lasers, when the modulation frequency was  in the range of 2-3 kHz as a result of the collisional relaxation processes in the discharge plasma. They found that the OG signal reappeared at higher modulation frequencies with an opposite phase. A better performance was achieved by modulating at a frequency in the range of 10- 20 kHz, with long term stability better than 5x10-9. The high frequency modulation makes possible rapid corrections to reduce the laser noise in the audio-frequency band.

      The stabilization of CO2 lasers by optogalvanic monitoring of the Doppler-profile has been found to be a convenient technique, when ultimate stability properties are not important. This is because of its low cost and simplicity when compared with more sophisticated methods such as saturation fluorescence. This method is particularly well suited for output equipment, because it does not require additional detectors to obtain the error signal. The optogalvanic signal could be greatly improved by picking up the signal from a low-pressure amplifier discharge [64]. 

Schneider et al. [65] locked the CO laser to the absorption Lamb dip detected by the OG technique in an external low-pressure CO discharge and were able to obtain a relative stabilization in the 100 kHz range. A third derivative technique was applied in order to avoid frequency shifts due to asymmetry in the line shape (see Fig. 20). This locking method has particular importance for metrology. Tsai et al. [66] have stabilized a CO2 laser at the center of the regular lines, using the rf optogalvanic Lamb dip. The frequency stability is estimated to be better than 100 kHz. The rf optogalvanic Lamb dip should be applicable to sequence-band CO2 lasers and CO and N2O lasers.

5.5 Optogalvanic spectroscopy in flames

 

Fig. 21 Experimental arrangement for optogalvanic spectroscopy in flames using a pulsed dye laser [Adapted from Ref # 49]

In contrast to electrically sustained discharges, the flame is a chemically sustained mild plasma. Any analytical flame can be used to investigate atoms, molecules and free radicals using the optogalvanic technique. In the experimental setup of Fig. 21 a pair of tungsten wires are placed symmetrically around an analytical flame produced by a slot burner. The tungsten wires act as the cathode while the grounded slot burner head acts as the anode. The high voltage power supply maintains the cathode at a negative potential of about 1 kV with respect to the anode. The tunable dye laser beam is focused inside the flame between the two symmetrically placed electrodes. By changing the wavelength of the laser beam the condition of resonance, with transitions in atomic or molecular species in the flame, is obtained. The laser-induced transitions between a pair of levels perturbs the thermal population and leads to the optogalvanic effect, as described earlier. The pulsed optogalvanic signals are amplified before entering the boxcar averager, which gets the reference signal of the corresponding laser pulse from the combination of the beam splitter and the photodiode. The boxcar averages the signal over 30 or more laser pulses to enhance the signal to noise ratio.

Analytical flame spectrometry has been used in conjunction with the optogalvanic effect to detect trace metallic elements [67, 68]. Turk et al. [67] were one of the first to apply the optogalvanic effect in enhancing the ionization of atomic species present in the flame to develop a hybrid technique of detecting atoms and molecules requiring both laser excitation and thermal ionization. Hastie and coworkers [68] have reviewed the success of optogalvanic spectroscopy in obtaining absorption spectra of atomic and molecular species in flames by alleviating the problems associated with  optical monitoring. Analytical flame spectrometry utilises the optogalvanic effect for trace metal detection with great promise for many metallic elements. It can also probe the ionization effects in flames, and ionization cross sections and ion mobilities may be determined from an analysis of OG signals.

5.6 Optogalvanic studies on mobility of ions and soot formation in flames

                               

Fig. 22 Schematic oscilloscope trace for irradiation of a C2H2/air flame containing Na and U atoms with a single laser pulse at 539.9 nm [Adapted from Ref # 69]

Smyth and his group [69, 70] have used a slightly modified experimental setup than that shown in Fig. 21, to measure the mobility of atomic ions and small particles. Laser enhanced ionization creates a pencil of ions along the path of the laser in the flame.The ions are collected by the electrode at a distance from the point of laser impact, and the time from their creation to the collection measured for a series of applied potentials. The observed OG signal results from the increased collisional ionization rate of the neutral atoms in an electronically excited state relative to that in the ground state. Sodium and Uranium were introduced into the flame as salt solutions aspirated into the burner.

The tunable pulsed dye laser is tuned to an electronic transition of the neutral metal atom and a typical OG signal is schematically shown in Fig. 22. The first feature of the trace is the electron signal that peaks at about 0.8 microsecond. This is followed by the ion signals due to the sodium ions arriving 10 microsecond after the laser pulse and the uranium ions arriving 21 microsecond after the laser pulse. These experiments  directly measure the velocity of metal ions in a flame that leads to the determination of the mobility of the ion. A comparison of the direct experimental determinations of ion mobilities, of Li, Na, K, Ca, Fe, Sr, Ba, In, Tl, and U, with the Langevin theory [71] shows substantial improvement over previous studies.

The chemistry of soot formation in flames is a complex field where the key precursors leading to the formation of carbon particles are not very well understood. There are fundamental processes involving the roles of ions and polycyclic aromatic compounds that need to be studied. Smyth and Mallard [70] have carried out OG spectroscopy of flames to elucidate the formation of small particles in a C2H2/air flame. It has been found that laser-induced ionization signals observed, at the sooting limit, may be attributed to rapid heating of small particles with subsequent thermal ionization. A rough estimate of the particle mass (2300 -6300 amu) and particle size (1.6- 2.2 nm) for the ionized species has been obtained. These results indicate that the OG signals are due to very small particles which appear early in the soot formation process. 

5.7 Studies on Rydberg states of atoms

The Rydberg formula was developed to describe energy levels of hydrogen atom but it can be used to describe highly excited states of a multi-electron  atom or molecule. Such states, called Rydberg states, converge on an ionic state with an ionization energy threshold associated with a particular ionic core configuration. Laser optogalvanic spectroscopy is a novel technique for the investigation of Rydberg states of atoms and molecules, with very large principal quantum numbers (n= 20 to n 100), in discharge cells. Low pressure glow discharge is characterized by high electron temperatures in the range of 104 to 106 K and has significant population density in the excited states particularly the long lived metastable states. Such metastable levels have been used as intermediate levels in two-photon excitation of Rydberg states [72].

Baig and coworkers have studied the highly excited odd-parity states of mercury using two-photon and two-step excitation from the metastable state using a rf discharge cell.  The 6s6p3P0 metastable state was used in the observation of the 6snp3P0(13 n 27), 6snp3P2(13 n 30) and 6snf3F2(10 n 42) Rydberg series [73]. The 6s6p3P2 metastable state was used in a two-step excitation via 6s7s3S1 intermediate level to record the 6snp3P0(10n18), 6snp3P1(10n41), 6snp3P2(10n70), and 6snp1P1(10n42) Rydberg series [74]. The corresponding even-parity Rydberg series 6sns3S1(13n50), 6snd1D2(6n18), 6snd3D1(6n14),  6snd3D2(6n15) and 6snd3D3(6n59)   have been reported in a subsequent publication [75] where the detection of parity forbidden transitions 6snp3P1(44n50) in the presence of krypton as the buffer gas in the dc discharge has been attributed to the increase in transition probability with increase of atomic weight of the buffer gas.

5.8 Optogalvanic spectroscopy of molecules

Iodine molecule (I2) has acquired the same role in electronic spectroscopy of molecules as does the hydrogen atom in atomic spectroscopy. Rettner et al. [76] were the first to investigate the optogalvanic spectrum of I2 and found that the B-X electronic system closely resembles its fluorescence excitation spectrum except when the laser beam is coincident with the discharge axis.  It was found that the OG signal changes sign in the dc discharge tube when moving from the cathode to the anode, suggesting that more than one mechanism is responsible for OG effect in iodine. Webster et al. [58] used a pulsed dye laser to probe the discharge in iodine vapour to study the optogalvanic spectrum with a view to investigate electron photodetachment from I- ions. It is found that the first photodetachment threshold for the production of I (2P3/2) atoms occurs at 405.180.02 nm leading to the electron affinity Ea(I)= 3.05910.0001 eV. Rai et al [77] have studied the Doppler limited OG spectrum of the B-X system of I2 molecule in the wavelength region 570 -630 nm. A weak continuum is found to be superposed over the discrete spectrum of the B-X system. When the applied voltage is increased, the OG signals corresponding to some of the discrete bands, in the lower wavelength region, change from positive to negative. The continuum OG signal in the entire region also  changes from positive to negative but the remaining spectrum is much less affected by the change in applied voltage. These studies indicate that there are multiple mechanisms responsible for the OG effect in iodine molecule.

Vasudev and Zare [78] have studied the OG spectrum of the A-X electronic system of HCO radical. Most of the levels of the excited state of HCO are diffuse and laser induced fluorescence (LIF) is inapplicable to investigate the A-X system due to extremely low fluorescence quantum yield of the A state. The OG spectrum was recorded, in an electrodeless rf discharge through acetaldehyde to produce HCO, by using rhodamine 6G dye laser in the 580-620 nm region. It is found that the OG effect is especially large when the laser excited transition involves a predissociated state. The analysis of the OG spectrum has been done by assuming a number of predissociation mechanisms.

5.9 Intracavity optogalvanic spectroscopy

Intracavity optogalvanic spectroscopy is an ultrasensitive laser-based technique for detection of 14C-labeled carbon dioxide [79, 80]. The strongest lasing transitions in 14CO2 laser are at 11.8 microns {P(20)} and at 11.3 microns {R(20)}, significantly longer in wavelength than the lasing transitions of the other stable isotope CO2 lasers.The experimental system also includes a small 12CO2 laser for normalization, which operates at 10.6 microns and the emission is in resonance with only 12CO2 molecules in the external reference cell. The analyte cell is placed inside the 14CO2 laser cavity. The partial transmittance of the front mirror of the 14CO2 laser makes it possible to allow 1.5 W of the 10 W 12CO2 laser beam to be incident onto the intracavity sample cell. Both lasers are modulated at different frequencies and 12CO2 and 14CO2 OG signals are acquired simultaneously by phase sensitive lock-in technique.

Each sample cell is capacitively coupled to a tunable, low power RF oscillator circuitry that initiates and sustains the glow discharge. Since glow discharge is generated via the RF voltage applied to the electrodes mounted outside the analyte cell, the electrodes do not come into direct contact with the sample being analyzed. Measuring the OG signals simultaneously for both cells and taking the double ratio of signals eliminates errors due to laser fluctuations. 12CO2 Normalization is achieved by the OG signal from the external cell, while the 14CO2 is normalized to the OG signal from the laser tube itself.

Using this ultra sensitive OG detection technique, limits of detection for 14C/12C ratios near 10-15 have been obtained. With a 15 W 14CO2 laser, a linear calibration  with samples from 5x10-15 to greater than 1.5x10-12 in ratios has been demonstrated [80]. Possible applications of this technique include microdosing studies in drug development, individualised sub-therapeutic tests of metabolism, carbon dating and real time monitoring of atmospheric radiocarbon. The most recent work using this technique has, however, shown serious problems of reproducibility in the detection of radiocarbon and this method is found as not a viable method [81]  

6. Future trends in optogalvanic detection

The very sensitive technique of optogalvanic spectroscopy is known to suffer from stability and reproducibility problems. Attempts are being made to get a deeper knowledge of the underlying physics to improve the performance of OGS. Hollow-cathode discharge is known as a reservoir of sputtered atoms and the analysis of OGS is based on the understanding that the sputtered atoms which are not returned back to the cathode are thermalized by the buffers in the cathode dark space. Zhechev and Steflekova [82] have studied the conductivity of an ensemble of sputtered atoms in a hollow cathode discharge as a function of their alignment and orientation.

        Berglund [83] has designed and constructed a miniature high-precision, isotope-resolving molecular spectrometer based on the optogalvanic effect. The most important component of this microsystem is a microplasma source in the form of a split ring resonator. The plasma sources developed in this work are the first ever miniature devices to be used in optogalvanic spectroscopy (OGS). These sources have been manufactured in both printed circuit board and alumina for compatibility with other devices in the spectrometer.

Preliminary work on the microsystem has confirmed that OGS scales well with miniaturization. The signal strength does not decrease as the volume is reduced and the stability and reproducibility are greatly increased. A major benefit of the miniature sample cell is the miniscule amount of sample it requires. The use of this instrument for exploration of planets and moons, is expected to help the detection of extraterrestrial life.

7. Conclusion

This chapter provides a historical overview of the Joshi effect, discovered between 1939 and 1940 in Professor S.S. Joshi's laboratory at Banaras Hindu University, India. This discovery emerged from Joshi's decade-long exploration of silent or electrodeless discharges, which affect gas-phase chemical reactions with high activation energies. In the mid-1920s, Joshi had  utilized the silent electric discharge in Professor F.G. Donnan's laboratory at University College, London, to study nitrous oxide decomposition.

The Joshi effect refers to the changes in discharge current—either positive or negative—within a high-voltage electrodeless AC corona discharge when external radiation strikes the discharge tube. This phenomenon is akin to the Optogalvanic effect, a term coined in 1976 to describe transient current or voltage changes in a stable DC discharge caused by a modulated laser beam. The Optogalvanic effect is often associated with the Penning effect, identified in 1928, which involves an increased discharge threshold potential in argon-neon mixtures when irradiated by external neon light. This effect occurs due to the destruction of metastable neon states, impacting secondary ionization processes during gas discharge. In contrast, the Joshi effect occurs after the discharge is established, and it is exclusively observed in electrodeless systems, making it a more general phenomenon closely related to the Optogalvanic effect.

Wagh and Deshpande [29] emphasize that understanding plasma formation is crucial to explaining the Joshi effect and its response to irradiation. They suggest viewing an ozonizer as a cylindrical capacitor, where the geometry influences plasma formation. This perspective elucidates many aspects of the Joshi effect, including the role of optical excitation.

Optogalvanic spectroscopy operates on the principle that photon-induced spectral transitions can either enhance or diminish the ionization rate of the discharge, resulting in negative (-ΔV) or positive (+ΔV) voltage signals. The OG spectrum of atomic and molecular species can be recorded by varying the wavelength of the excitation radiation. This straightforward technique can be applied to various plasma sources, including positive column DC discharges, hollow-cathode DC discharges, radio-frequency discharges, and flames.

The OG effect manifests as voltage changes across a DC glow discharge tube due to illumination, particularly with resonant light absorbed by atomic species in the plasma. In pulsed experiments, the light is briefly activated (10⁻⁸ seconds), and the discharge voltage is recorded without light for many microseconds. The temporal evolution of the OG signal, ΔV(t), can be described through a semi-empirical theory. Typically, the OG signal is initially negative unless the population of the involved energy levels is inverted. The initial negativity flips to positivity, especially for transitions originating in metastable levels, where the upper state's radiative relaxation depletes the metastable state, resulting in a positive ΔV to balance the decreased electron multiplication in the discharge.

Several key applications of OG spectroscopy illustrate its capacity to elucidate a wide range of atomic and molecular processes. This technique effectively detects weak absorptions in gaseous discharges and allows the study of high-lying states, including Rydberg states, through radiative transitions from well-populated metastable levels. Combining Doppler-free spectroscopy with optogalvanic detection has led to sensitive new high-resolution methods. Additionally, Laser-induced OG spectroscopy provides an accurate, economical way to calibrate the wavelengths of tunable lasers and offers active frequency stabilization for various laser types. It has also been employed to determine the photodetachment thresholds of negative ions in atomic and molecular systems.

While comprehensive review articles and conference proceedings have addressed recent developments in this analytical technique, there is a notable scarcity of books consolidating all aspects of OG spectroscopy. The only existing book on this topic [84] was published in Russian in 1991, with an English translation released in 1999. This publication aims to fill that gap and benefit students and young scientists interested in novel spectroscopic techniques.

      

References

1. S.S. Joshi, Threshold potentials and reactivity under electrical discharge, Curr. Sci. 8 (1939) 548

2. M.I. Abu-Taha, F.A. Zhour, Simultaneous intra-cavity detection of some physical effects in a CO2 laser, Int. J. Natu. Engg. Sci. 3 (2009) 26

3. J.S. Townsend, The conductivity produced in gases by the aid of ultra-violet light, Phil. Mag. 3 (1902) 557

4. G. Moeller, J. Rigden,  Observation of laser action in the R-branch of CO2 and N2O vibrational spectra,  Appl. Phys. Lett. 8 (1966) 69                 

5. J.E.M. Goldsmith and J.E. Lawler,   Optogalvanic spectroscopy, Contemp. Phys. 22 (1981) 235

6. F.M. Penning, Demonstração de um novo efeito fotoelétrico, Physica 8 (1928) 137

7. F.M. Penning, Starting potentials of the corona discharge in neon,   Phil. Mag. 11 (1931) 961

8. F.M. Penning, Nieuwe metingen over de doorslagspanning van eldelgassen, Physica 12 (1932) 66

9. M.J. Druvedeya and F.M. Penning, The mechanis of electrical discharges in gases of low pressure,  Rev. Mod. Phys. 12 (1940) 87

10. M.J. Druvedeya and F.M. Penning, Errata: The mechanis of electrical discharges in gases of low pressure, Rev. Mod. Phys. 13 (1941)  72

11. S.S. Joshi and V. Narasimhan, Light effect in chlorine under electric discharge,   Curr. Sci. 9 (1940) 535

12. H.J. Arnikar, From the Joshi effect to Optogalvanic spectroscopy, Curr. Sci 51 (1982) 1103

13. G.W. Series, Laser spectroscopy, In:  Laser Spectroscopy and other topics, Ind. Acad. Sci. Bangalore (1985) 568

14. S.S. Joshi and P.G. Deo, A light effect in chlorine under electric discharge; Influence of the intensity and frequency,  Nature 151 (1943) 561

15. S.S. Joshi, G.A. Elliot, and R.W. Lunt, The decomposition of nitrous oxidein the silent electric discharge, Trans. Farad. Soc. 23 (1927) 227

16. G.S. Deshmukh and Y.D. Kane, Interaction of nitrous oxide, nitrogen and phosphorus under silent electric discharge, Ind. Acad. Sci (1946) p. 247-55        

17. S.S. Joshi, Further consequences of the activated layer postulate in the mechanism  of the light-effect, Curr. Sci  16 (1947) 19     

18. S.N. Tewari, Study of the Joshi effect in iodine vapor in the presence of powdered  iodine under electrical discharge,  Ind. J. Phys. 22 (1948) 553

19. A.P. Saxena, Potential reversibility of Joshi effect in iodine vapor under silent electric discharge,  J. Ind. Chem. Soc. 28 (1951)  239    

20. S.R. Khastgir and P.S.V. Setty, Positive and negative Joshi effect in ac silent discharges in  iodine vapour, Proc. Phys. Soc.(London)   B 65 (1952) 823                                                        

21. S.R. Khastgir and P.S.V. Setty, Some observations on the effect of light on discharge through iodine vapor,  Proc. Nat. Inst. Sci. (India) 19 (1953) 631

22. S.S. Joshi and G.S. Deshmukh, An effect of light in gases under electric discharge  Nature 147 (1941) 806

23. P.G. Deo, The new light-effect in chlorine under electrical discharge,  Proc. Ind. Acad. Sci. A19 (1944) 117

24. R. Parshad and S.J. Karim, Decrease of an electrical discharge by external radiation,  J. Chem. Phys. 17 (1949) 667           

25. W.L. Harries and A. von Engel, The mechanism of the low frequency electrode-less discharge and the influence of irradiation, Proc. Phys. Soc. B 64 (1951) 916

26. W.L. Harries and A von Engel, Influence of irradiation on the low frequency electrode-less discharge in chlorine, J. Chem. Phys. 19 (1951) 514                          

27. R. Parshad, Explanation of the  Joshi effect, Nature 155 (1945) 362                 

28. A.K. Saha and M. Ghosh, The mechanism of the low frequency electrode-less discharge in chlorine and the influence of irradiation, Proc. Phys. Soc. B 65 (1952) 304  

29. S.M. Wagh and D.A. Deshpande,  Revisiting the Joshi effect, Curr. Sci. 101 (2011) 1182

30. M. Venugopalan,  Photosensitivity ofelectrical discharges in gases,  Nagpur Univ. J. 2 (1971) 22      

31. V.G. Kher and M.G. Kelkar, High frequency electrodeless  discharge and the influence of irradiation: Part I,   J. Sci. Ind. Res. B21 (1962) 293                    

32. V.G. Kher and M.G. Kelkar, High frequency electrodeless  discharge and the influence of irradiation: Part II,  J. Sci. Ind. Res.  B21 (1962) 360                   

33. V.G. Kher, Mechanism of a low frequency (10 Hz to 10 kHz) ac electrodeless discharge & Joshi effect,  Nagpur Univ. J.  1 (1971) 96                  

34. L.B. Loeb, Ion-ion recombination theory over an extended pressure range  Nagpur Univ. J.  2  (1971) 1    

35. S.S. Joshi,The decomposition of nitrous oxide in the silent electric discharge, Trans. Faraday Soc.  25 (1929) 108;  118;  143      

36. M. Marklund and P.K. Shukla, Nonlinear collective effects in photon-photon and photon-plasma interactions,    Rev. Mod. Phys.. 78 (2006) 591     

37. C. Kenty, Role of metastable (3P2) Hg atoms   in low current discharges in Hg rare gas mixtures, Phys. Rev. 80 (1950) 95. 

38. K.W. Meissner and W.F. Miller, Influence of irradiation on the characteristic of a glow discharge  in pure rare gases, Phys. Rev. 92 (1953) 896 

39. M.G. Drouet and J.P. Novak, Effect of illumination of a neon discharge on the electron energy distribution, Phys. Lett. A 34 (1971) 199

40. N.S. Kopeika, J. Rosenbaum and R. Kastner, Abnormal glow discharge of visible radiation, Appl. Opt. 15 (1976) 1610

41. R.B. Green, R.A. Keller, G.G. Luther, P.K. Schenck, and J.C. Travis, Galvanic detection of optical absorptions in a gas discharge, Appl. Phys. Lett. 29 (1976) 727

42. R.B. Green, R.A. Keller,  P .K. Schenck, J.C. Travis, and   G..G. Luther, Optogalvanic detection of species in flames, J. Am. Chem. Soc. 98 (1976)  8517

43. R.B. Green, R.A. Keller, G.G. Luther, P.K. Schenck, and J.C. Travis, Use of an optogalvanic effect to frequency-lock a continuous wave dye laser, IEEE J. Quant. Electron. 13 (1977) 63

44. D.S. King, P.K. Schenck, K C. Smyth and J.C. Travis, Direct calibration of laser wavelength and bandwidth using the optogalvanic effect in hollow cathode lamps, Appl. Opt. 16  (1977) 2617

45. K.C. Smyth and P.K. Schenck, Optogalvanic spectroscopy of a neon discharge: Mechanism studies, Chem. Phys. Lett. 55 (1978) 466                  

46. K.C. Smyth, R.A. Keller and F.F. Crim,  Photon-induced ionization changes in a neon discharge,  Chem. Phys. Lett. 55 (1978) 473                        

47. P.K. Schenck, W.G. Mallard, J.C. Travis and K.C. Smyth, Absorptionj spectra of metal oxides using optogalvanic spectroscopy,  J. Chem. Phys. 69 (1978) 5147    

48. G. Erez, S. Lavi and E. Miron,  A simplified theory of the optogalvanic effect,  IEEE J. Quantum Electron QE-15 (1979) 1328    

49. S.N. Thakur, Optogalvanic spectroscopy and its applications, In: Molecular and Laser Spectroscopy- Advanced and Applications, Volume 2, V.P. Gupta and Yukihiro Ozaki (eds),  Elsevier (2020) pp 217-244    

50. S. Levesque, F. Babin, and J.M. Gagne,  Calibration of pulsed laser wavelength by optogalvanic spectroscopy in hollow cathode discharges,  IEEE Trans. Instrum. Measurement 42 (1993) 251  

51. C.R. Webster, Optogalvanic wavelength calibration for laser monitoring of reactive atmospheric species,  Appl. Optics 21 (1982) 2298    

52. J E. Lawler, A.I. Ferguson, J.E.M. Goldsmith, D.J. Jackson, and A. L. Schawlow, Phys. Rev. Lett. 42 (1979) 1046

53. P. Camus, Atomic spectroscopy withoptogalvanic detection, J. de Physique. C7 (1981) 87

54. T.W. Hansch, D.R. Lyons, A.L. Schawlow, A. Siegel, Z-Y Wang, G-Y Yan, Polarization intermodulated excitation (POLINEX) spectroscopy of helium and neon, Optics Commun, 38 (1981) 47

55. B. Barbieri, N. Beverini, and A. Sasso, Optogalvanic spectroscopy, Rev. Mod. Phys. 62  (1990)  603

56. K. Kawakita, K. Fukuda, Y. Adachi, and C. Hirose, Doppler-free optogalvanic spectrum of He2  b3g- f3u transition, J. Chem. Phys. 82 (1985)  653

57. I.S.  McDermid, and C.R. Webster,  Optogalvanic photodetachment spectroscopy, J. de Physique  C7  (1983) 461          

58. C.R. Webster, I.S. McDermid, and C.T. Rettner, Laser optogalvanic photodetachment spectroscopy: A new technique for studying photodetachment thresholds with application to I-,  J. Chem. Phys. 78 (1983) 646

59. R. Klein, R.P. McGinnis, and S.R. Leone, Photodetachment threshold of CN- by laser optogalvanic spectroscopy,  Chem. Phys. Lett. 100 (1983) 475

60. M.G. Tonzeau, J. Gousset, D. Jolly, et al. , Spectroscopy and kinetics of an oxygen glow discharge: In Nonequilibrium process in partially ionized gases, M. Capitelli and J.N. Beardsley (eds.) Plenum Press, New York (1990) pp 571-80

61. W.H. Thomason and D.C. Elbers, An inexpensive method to stabilize the frequjency of a CO2 laser,   Rev. Sci. Instrum. 46 (1975) 409 

62. A.L.S. Smith and S. Moffatt, Optogalvanic stabilized CO2 laser  Opt. Commun.  30  (1979)  213

63. S. Moffatt and A.L.S. Smith, Temperature perturbation model of the optogalvanic effect in CO2 laser discharges,  J. Phys. D 17  (1984) 59

64. G.L. Bourdet and P. Houmault, Optogalvanic effect in CO2 amplifier, Opt. Commun.   58  (1986) 331

65. M. Schneider, A. Hinz, A. Groh et al., CO laser stabilization using the optogalvanic Lamb dip,  Appl. Phys. B 44 (1987) 241

66. C.-C. Tsai, T. Lin, C-Y. Shieh, et al. CO2 laser frequency stabilization using the radio-frequency optogalvanic Lamb dip, 

67. G.C. Turk, J.C. Travis, J.R. DeVoe, and T.C. O’Haver,  Analytical flame spectrometry with laser enhanced ionization, Anal. Chem. 50 (1978) 617

68. P.K. Hastie and J.W. Hastie, Optogalvanic spectroscopy-Application to combustion systems, Opt. Engg. 20 (1981) 204522 

69. W.G. Mallard and K.C. Smyth, Mobility measurements of atomic ions in flames using laser-enhanced ionization,  Combustion and Flames 44 (1982) 61

70. KC. Smyth and W.G. Mallard, Laser-induced ionization and mobility measurements of very small particles in premixed flames at the sooting limit, Combustion Science and Technology 26 (1981) 35

71. P. Langevin, A fundamental formula of kinetic theory, Ann. Chim. Phys. 5 (1905) 245

72. H. Rinneberg and J. Neukammer,  High resolution laser spectroscopy of odd-parity barium Rydberg states populated by forbidden atomic transitions,  J. Physique Colloque 44 (1983) C7-177

73. M.A. Zia, B. Suleman and M.A. Baig, Two-photon laser optogalvanic spectroscopy of the Rydberg states of mercury by rf discharge,  J. Phys. B 36 (2003) 4631

74. M.A. Zia and M.A. Baig, Two-step laser optogalvanic spectroscopy of the odd-parity Rydberg states of atomic mercury,  Eur. Phys. J. D 28 (2004) 323

75. M.A. Zia and M.A. Baig, Laser optogalvanic spectroscopy of the even-parity Rydberg states of atomic mercury,  J. Opt. Soc. Am. B 22 (2005) 2702

76. C.T. Rettner, C.R. Webster, R.N. Zare, Laser optogalvanic effect in a pure iodine discharge,  J. Chem. Phys. 85 (1981) 1105

77. A.K. Rai, S.B. Rai, S.N. Thakur and D.K. Rai, Effect of DC voltage on the laser optogalvanic spectrum of iodine molecule, Chem. Phys. Lett. 138 (1987) 215

78. R. Vasudev and R.N. Zare, Laser optogalvanic study of HCO A~ state predisposition, J. Chem. Phys. 76 (1982) 5267

79. D.E. Murnick, O. Dogru and E. Ilkmen, Intracavity optogalvanic spectroscopy: An analytical technique for 14C analysis with subattomole sensitivity, Anal. Chem. 80  (2008)  4820

80. D.E. Murnick, O. Dogru, E. Ilkmen, Nuclear Instruments and Methods in Physics Research B: Beam Interactions with Materials and Atoms. 268 (2010) 708

81. A. Persson, M. Salehpour, Nuclear Instruments and Methods in Physics Research B: Beam Interactions with Materials and Atoms. 361 (2015) 8

82. D. Zhechev and V. Steflekova, Optogalvanic effect in degenerate magnetic states of sputtered atoms in a glow discharge,  Journal of Physics: Conference Series 558 (2014) 012032

83. M. Berglund, Miniature plasma sources for high-precision molecular spectroscopy in planetary exploration,  Doctoral Thesis (summary), Uppsala University (2015)

84. V.N. Ochkin, N.G. Preobrazhensky, N.Y. Shaparev, Optogalvanic Effect in Ionized Gas, Lebedev Physical Institute University Press, Moscow (1999)