NDT.net • Sep 2005 • Vol. 10 No.9

On the transport of Carbon in CdS by photoacoustics

A.K. Balasubramanian,
Department of Physics, Sourashtra College, Madurai 625004.
School of Physics, Madurai Kamaraj university, Madurai 625 021
N.Sankar and K.Ramachandran
Department of Physics, Yadava College, Madurai 625014.
School of Physics, Madurai Kamaraj university, Madurai 625 021.

*Corresponding Author Contact:
Email: thirumalchandran@yahoo.com, Internet:


Temperature dependent phase transformation in CdS is studied by a non destructive procedure namely photoacoustics. Single crystals of CdS have been grown by physical vapour transport technique.The transport of carbon in CdS is studied here by photoacoustic spectroscopy. An interesting transition to CdCO3 is observed at about 800C which has been discussed in detail with reference to the results available in literature.

Keywords: NDT, Photoacoustic spectroscopy (PAS), CdS, Thermal diffusivity.


The photoacoustic technique(PA) has been effectively used to characterize the materials because of its great versality as a non-destructive and non-invasive methods for the evaluation of materials. Though it is used for studying the optical properties as well as the thermal properties, the application of this photoacoustic spectroscopy as a non-destructive method is very rare. Here the transport of carbon in CdS is studied by photoacoustics in a non destructive way which is unconventional.

The following figure explains the photoacoustic technique and represent the block diagram for the photoacoustic experimental set up.

Fig.2 Schematic diagram of the photoacoustic(PA) spectrometer.

When a modulated light from the chopper is absorbed by the sample located in a sealed PA cell, the non-radiative decay of the absorbed light produces a modulated transfer of heat to the surface of the sample. This modulated thermal gradient produces pressure waves in the gas inside the PA cell that can be detected by the microphone (MIC) attached to the cell. The resulting acoustic signal not only depends on the amount of heat generated in the sample but also on how the heat diffuses through the sample It is then amplified by the lock in amplifier and the amplitude and phase are measured. The quantity , which measures the rate of heat diffusion in a material is the thermal diffusivity which is a unique parameter. The thermal properties of the sample are studied from this parameter.

Cadmium sulphide is a technologically important compound semiconductor used for sensors, solar cells, luminescence etc., and lot of experimental works related to these are already available in literature. Eventhough, Knudson and Gupta [1] as early as in 1998 have reported an interesting transformation kinetics for the shock wave induced phase transition in CdS,where transformation to metastable state was observed for a pressure of about 60 to 70 kbar for a axis, recently Portillo etal [2] have reported another work on the phase transition from CdS to CdCO3 ( hexagonal). When thin films of CdCO3 was attempted (and in fact this happens to be the first ever report on the growth of cadmium carbonate thin film) by chemical bath (CB) method, they have observed that between 23 -800 °C, there will be an intermediate phase between CdS and CdCO3 and at 800 °C , the zinc blende CdS is exclusively seen. ie CdCO3 was predominant in the layer at 230C whereas it is CdS at 800C. ie a phase transformation was observed at temperature of 800C where zinc blende CdS becomes stable. This is an interesting case of phase transition . This transition is identified from the physical properties of the film by means of XRD and optical absorption methods.The energy band gap of this direct transition was also worked out to confirm this transition.

For systems like CdS or CdCO3 the melting points are so high that a transition at 800C is quite unusual. So this needs a detailed investigation to find whether S is replaced by CO3.Here we report our study on CdS crystal in the presence of carbon by photoacoustic (PA) measurements which has successfully been applied to the non destructive way of testing the sample.

Photo acoustics(PA) :

Physical vapour transport technique is used to grow the CdS crystals and the details of the growth are already given[3]. Good single bulk crystals in cm size are obtained with high purity. Crystalline nature and electrical properties are also already reported[3]. Wafers of thickness less than 1 mm are sliced from the as grown samples and polished with 10% of bromine in ethyl alcohol for about 2 minutes. The crystals are cleaved along the (110) faces with the thickness of the sample in few micrometres. These are used for the PA measurement. The present PA spectrometer used for the study of phase transition is set up with a Xenon lamp (450 Watts), monochromator (100-1000nm, Jobin Yvon), mechanical chopper ( PAR ), photoacoustic cell, a condenser mike, a lock in amplifier (PAR DSP 7225) for amplifying the PA signal by which the amplitude and phase of the signal are measured.

We have fabricated the heating compartment using nichrome (28 gage) wire as heating element fixed with plaster of paris around the PA cell to vary the temperature of the sample. The PA cell is made up of stainless steel. It is present inside the cylindrical container of length about 0.5 metre.The microphone compartment carrying the condenser mike is attached to the container by another rod of very narrow diameter. Thus it is isolated from the heating element to avoid direct heating of the condenser mike which detects the PA signal. The temperature can be varied up to 3000C and Al-Cr thermocouple is used to measure the temperature . This is digitally observed in 0C with an accuracy of 10C. This set up is first calibrated with an accurate thermometer for different temperatures. Then only the PA measurements for the CdS system for various temperatures are observed.

The crystal is then introduced in the PA cell and very close to the sample a piece of carbon is placed as a backing material. In general, when PA measurements are done, the thermal properties of the backing material should not affect the thermal properties of the sample. But here , we have deliberately used carbon as the backing material so that on heating whether there is any possibility for the carbon atoms to get transported into CdS to form CdCO3(as there is always oxygen present in the PA cell). The temperature is varied from the ambient to the maximum of 1200C in steps of 50 °C.


The beam of white light from Xenon lamp is chopped at an audio frequency of 22 Hz. This frequency is chosen here from the previous works on CdS crystal [3] and is the characteristic frequency of the sample. It is then allowed to fall on the sample which is mounted inside the airtight PA cell. The periodic absorption of energy by the sample results in the heating of the air medium and generates a sound wave at the chopping frequency. This is detected by the mike and fed to the digital storage oscilloscope (DSO GOULD 20 MHz ) and lock in amplifier. The variation of PA amplitude with the temperature has been measured and recorded as shown in Fig.1a. This is repeated again for two more chopped frequencies of 35 Hz and 55 Hz. Similarly the variation of phase with temperature also has been measured and recorded as shown in Fig.1b. In both the observations (Fig 1a &1b), the PA amplitude first increases with the temperature and starts to decrease at a temperature around 800C.

Fig 1a. Temperature Vs PA signal for CdS.

Fig 1b. Temperature Vs Phase

The reason for this was analysed to find transport of carbon into the sample CdS to form CdCO3 due to the oxygen present in the PA cell as the CdS crystal grown is enriched with Cd concentration. As we have mentioned earlier, this possibility is rare when the melting points of CdS or CdCO3 are considered.

But Shivrin etal [5] have reported some interesting results as early as in 1989 when they studied the mechanism of oxidation of metal sulphides like CdS, ZnS,In2S2 etc by water vapour method and have found that the oxidation begins at a temperature 0.3285 Tb where Tb is the boiling point of metal in metal sulphide. The boiling point of metal ie Cd in CdS is 1050 K and so the oxidation at about 350 K (850C) is expected from the empirical relation proposed by Shivrin etal.

Now, our measurements also show a drastic change at 850C. Therefore the formation of CdCO3 cannot be completely ruled out at intermediate temperatures.This decreases the PA amplitude after 800C showing a transition from CdS to CdCO3 and thus the phase of CdCO3 is maintained at 800C. This has been further verified from the variation of phase measurements with temperature as shown in Fig.1b.

Here again, the phase of the signal is found to increase first and at a particular temperature of 800C, it remained constant. Thereafter it starts to decrease as the temperature increases. This shows that this is the temperature at which the zinc blende phase of CdS is maintained from the hexagonal (rhombohedral) phase of CdCO3. By this way, we have demonstrated the Xray diffraction measurements of Portillo etal proposing the growth of CdCO3 from CdS. ie For example,

Cd2+ + O2 --------> 2CdO (at 800C)

Since the carbon is the backing material,

2CdO + 2O2 +3C --------> 2CdCO3 + °C

So, there is a possibility to going from CdS to CdCO3 and this CdCO3 is not very stable. 3vice versa

In general, the PA signal is proportional to ( µ/T)(2rCs) i.e it is inversely proportional to the temperature. Now if we look at Fig:1a, we see there is an increase first and then only at about 800 °C there is a decrease as expected. Thermal diffusivity is also worked out from PA measurements for various temperatures in this region and the variation of diffusivity as a function of temperature is shown in Fig.1d. It increases first and at around 800C, it starts to decrease which shows that there is a change in the thermal property following the theory of Rosencwaig [4]. The characteristic frequency fch of CdS at 800C is 39 Hz and the thermal diffusivity is increased atleast by a factor of 3 compared to the ambient value as shown in the Table 1. This is also an useful result as the thermal properties are affected in any transition.

Fig 1c. Frequency Vs PA signal.

Fig 1d. Temperature Vs Thermal diffusivity

Results and discussion:

Portillo etal [2] have actually synthesized CdS by chemical bath method by taking CdCl2, KOH, NH4NO3 and SC(NH2)2 solutions mixed at equal volume proportion to have a pH of 8.3. When this solution is heated from ambient to about 1200C at different temperatures and deposited on a glass substrate, different samples are obtained. For these samples the XRD and optical absorption spectra measurements are made to find that below 800C, it is a mixed phase of CdS and CdCO3 and above 800C it is purely CdS.

In the present experiment, samples are not synthesized by chemical route whereas an indirect chemical reaction is initiated to obtain CdCO3 from CdS. So there is no one to one correspondence between this work and Portillo etal. Still we consider Portillo's work, as this is the only one work reported so far for CdS-CdCO3 transition. The present PA measurements and that of Portillo et al reveal one important point that there is a transformation at about 800C. Eventhough our PA measurements agree with the Portillo etal,it is interesting to see such a transformation at this temperature, as the melting point of Cd and C are very high. This is now explained here on the basis of the work of Shivrin etal.

Shivrin etal [5] have reported that the composition of reaction products was dependent on temperature and the H atom in the water vapour can combine (reduction) with the sulphur in CdS and this has also been observed but at lower temperature. ie. it is clear from this experiment, oxidation is possible in CdS at temperature ≥ 0.3285 Tb and reduction at low temperature. In the present PA experiment we have CdS and carbon apart from the air medium that is present in the PA cell. Therefore if we extend the concept of Shivrin etal, here the oxygen is available(from the air medium) to undergo oxidation process, at around 850C.

The present PA measurement shows this possibility of transport of carbon at temperature above 0.33 Tb approximately. ie. The CdS backed with carbon black when oxidized there is a possibility of CdCO3 at 800C. We expect CdS and carbon exist separately at lower temperature below 800C. This is also seen in Fig.1d where the variation of the PA signal with chopping frequency for very close temperatures is shown. We found unusual trend at 800C in the observations, as there is a cross over at 800C in this Fig.1c with the other temperature of 750C. The other curves at 70,75 and 85 all show uniform trend except this at 800. As there is an intersection between 750C and 800C, the mixed phase at this temperature is visually seen. .

Temperature (°C) Thermal diffusivity (10-7m2/s) CdS (10-7m2/s)
70 9.69 Ref
75 9.91
80 9.97 at 300 K
85 9.83 3.57
90 9.76
95 9.58
100 9.70
105 9.96
Table.1 Variation of thermal diffusivity with temperature

Similarly in Table.1, as the temperature increases from 700C there is an increase in thermal diffusivity whereas at 800C, this value begins to decrease. This means that there is an additional mass transport has taken place in CdS. But Portillo etal started with CdCO3 and CdS as two separate layers sandwiched and they have found the transition at 800C, going completely into CdS phase. ie to start with there is a mixed phase of CdCO3 and CdS but at 800C, there is only one phase of CdS. Here we have not started with two layers of CdCO3 and CdS but kept carbon very close to CdS with oxygen present. There is an unusual behaviour at 800C in our present case also, eventhough it is not a permanent change. It is reversible here also as seen by Portillo etal. ie during the fall of temperature, thermal diffusivity decreases to around 3.6x 10-7m2/s at room temperature, which is nothing but the value of thermal diffusivity of CdS at room temperature. ie CdS and CdCO3 can be seen in our present experiment also. This way the concept of non destructive testing is further verified.

This photoacoustic measurements could reveal the possibility of mass transport in CdS which would otherwise be difficult to measure by other conventional techniques.Here thermal diffusivity of CdS under various temperatures has been worked out from PA measurements and the change in this thermal property at 850C clearly indicates the transition in the presence of carbon.. The value (9.97x10-7m2/s ) is also not that of bulk CdS (which is 3.6x10-7m2/s) and so the diffusivity is essentially due to the presence of carbon and oxygen.


  1. M. D. Knudson and Y. M. Gupta, J. Appl.Phys, 91(12) 9561-9571, (1998)
  2. O. Portillo-Moreno, R. Lozada-Morales and O. Zelaya-Angel, Superficies y Vac., 15, 19-21, (2002)
  3. N. Sankar, K. Ramachandran and C. Sanjeeviraja, J.Crys.Growth, 235, 195, (2002)
  4. A. Rosencwaig, Photoacoustics and Photoacoustic Spectroscopy, Wiley, NY, (1980)
  5. G. N. Shivrin, V.A. Vostrikov, I. I. Smirnov and V.G. Kovgan, Izv Vyssh, Uchebn Zaved. Tsvetn, Metall. 50-4, Russ,(1989)
  6. Chemical Abstract 7440-43-9, Cadmium, Hand book of Chemistry and Physics 78th Edition, (1997-1998)
  7. D. D. Wagman, W. H. Evans,V. B. Parker, J.Phys.Chem.Ref.Data 11 [suppl.2] 1-392, 1982
  8. Lawless and Devries, J. Phys. Chem. Solids 25, 1119, (1964)
  9. O. Zelaya-Angel and R. Lozada -Morales, Phys.Rev. B 62, 13064, (2001)

© NDT.net |Top|