RESEARCH COMMUNICATIONS

CURRENT SCIENCE, VOL. 87, NO. 7, 10 OCTOBER 2004 981

Roth, S.), AIC Conference Proceedings 486, pp. 175–178. 13. Debnath, R. and Sahoo, R., UV–VIS–NIR, IR and mass spectral

characterization of some bismuth addition complexes of C60, syn-thesized through high temperature solid state reaction. Chem. Phys., 2004 (communicated).

14. Sahoo, R. and Debnath, R., Long-lived photoinduced charge sepa-ration in (Zn, lead) phosphate glass–C60 composites. Adv. Mater., 2003, 15, 287.

15. Evans, D. R., Fackler, N. L. P., Xie, Z., Rickard, C. E. F., Boyd, Peter D. W. and Reed, C. A., π-Arene/cation structure and bind-ing. Solvation versus ligand binding in iron (III) tetraphenylpor-phyrin complexes of benzene toluene, p-xylene and [60] fullerene. J. Am. Chem. Soc., 1999, 121, 8466.

16. Zheng, J. et al., Cyclic dimers of metalloporphyrins as tunable hosts for fullerene: A remarkable effect of rhodium. Angew. Chem. Int. Ed. Engl., 2001, 40, 1859.

17. Teng, B. et al., Growth and investigation of a new nonlinear optical crystal: bismuth borate BiB3O6. J. Cryst. Growth, 2001, 224, 280.

18. Iwasa, Y. et al., New phases of C60 synthesized at high pressure. Science, 1994, 264, 1570.

19. Dresselhaus, M. S., Dresselhaus, G. and Eklund, P. C., Science of Fullerene and Carbon Nanotubes. Academic Press, 1996.

20. Komatsu, K., Wang, G. W., Murata, Y., Tanaka, T., Fujiwara, K., Yamamoto, K., Sounders, M., Mechanical synthesis and charac-terisation of fullerene dimer C120. J. Org. Chem., 1998, 63, 9358.

21. Kunitake, M., Uemura, S., Ito, O., Fuziwara, K., Murata, Y. and Kamatsu, K., Strucutral analysis of C60 trimer by direct observa-tion with scanning tunnelling microscopy, Angew. Chem. Int. Ed. Engl., 2002, 41, 969.

22. Stenz, D., Blair, S., Goater, C., Feller S. A. and Affatigato, M., An investigation of bismuth borate glass structure using laser photo-ionization time-of-flight mass spectroscopy. Phys. Chem. Glasses, 2000, 41, 259.

23. Greenwood, N. N. Earnshaw, A., Chemistry of the Elements, Pergamon Press, Oxford, 1989, pp. 229–230.

24. Egorysheva, A. V., Burkov, V. I., Gorelick, V. S., Kargin, Yu. F., Koltashev, V. V. and Plotnishenko, B. G., Raman scattering in monocrystal BiB5O12. Phys. Sol. Stat., 2001, 43, 1655.

25. Weir, C. E. and Schroeder, R. A., Infrared spectra of the crystalline inorganic borates. J. Res. NBS, A, 1964, 68, 465.

26. Stone, C. E. et al., Structure of bismuth borate glasses. Phys. Chem. Glasses, 2000, 41, 409.

27. Bishay, A. and Maghrabi, C., Properties of bismuth glasses in relation to structure. Phys. Chem. Glasses, 1969, 10, 1.

28. Jun, L., Shuping, X. and Shiang, G., FT–IR and Raman spectroscopic study of hydrated borates. Spectrochim. Acta, A, 1995, 51, 519.

29. Gattef, E., Dimitrov, V. V., Dimitriev, Y. B. and Wright, A. C., Borate Glasses, Crystals and Melts, (eds Wright, A. C., Feller, S. A. and Hannan, A. C.), Society of Glass Technology, Sheffield, 1997, pp. 112–119.

30. Frolich, V. R., Bohaty, L. and Liebertz, J., The crystal structure of bismuth borate, BiB3O6. Acta, Crystallogr. Sect. C, 1984, 40, 343.

31. Filatov, S., Shepelev, Y., Bobnova, R., Sennova, N., Egorysheva, A. V. and Kargin, Y. F., The study of BiB5O12. J. Solid State Chem., 2004, 177, 515–522.

32. Rao, A. M., Eclund, P. C., Hodeau, J. L., Marques, L. and Nunez-Rigueiro, M., Infrared and Raman studies of pressured polymerized C60: Phys. Rev. B, 1997, 55, 4766.

33. Szymanski, H. A., Interpreted IR Spectra, Plenum Press, 1967, vol. 3, pp. 169–170; p. 195.

34. Debnath, R. and Sahoo, R., A synergistic composition for preparation of high concentration fullerene (C60)–glass composite in bulk monolith and their applications as nonlinear optical medium and optical limiter, etc. PCT Patent filing no. PCT/IB03/047, India, USA, UK, Germany and France.

35. Winkler, K., Noworyta, K., Kunter, W. and Balch, A. L., Study of redox active C60/Pd films by simultaneous cyclic voltammetry, and piezoelectric microgravimetry at an electrochemical quartz crystal microbalance. J. Electrochem. Soc., 2000, 47, 2097.

ACKNOWLEDGEMENTS. Financial support from DST and from CSIR, New Delhi, under CMM-0022 network programme is acknow-ledged. We thank the referees for their valuable comments. Received 6 March 2004; revised accepted 14 June 2004

Pyrolysis of rice husk

Anshu Bharadwaj1,*, Y. Wang2, S. Sridhar2 and V. S. Arunachalam2 1Center for Energy and Environment Studies and 2Department of Material Science and Engineering, Carnegie Mellon University, Pittsburgh, PA 15213, USA

Rice husk has proved to be a difficult fuel for gasi-fication and fluidized bed combustion because of the high ash content resulting in carbon conversion ineffi-ciency. We report the results of pyrolysis experiments of individual rice husk particles under controlled atmospheres of the confocal scanning laser micro-scope followed by SEM observations of the pyrolized structure of the particles. The results suggest a prefe-rential shrinkage of the particles in the transverse direction, implying the presence of strong and inert silica layer that retains its structure despite high temperatures. The cellulose and lignin component was preferentially consumed in geometrically arranged pores and channels. The role of silica appears to be more than just a geometric shield and thus complete carbon conversion may not be achievable.

RICE husk is the outer covering of paddy and accounts for 20–25% of its weight1. It is removed during rice mill-ing and is used mainly as fuel for heating in Indian homes and industries. Its heating value of 13–15 MJ/kg1,2 is lower than most woody biomass fuels. However, it is exten-sively used in rural India because of its widespread avail-ability and relatively low cost. The annual generation of rice husk in India is 18–22 million tons3 and this corres-ponds to a power generation potential of 1200 MW4. A few rice husk-based power plants with capacities between 1 and 10 MW are already in operation and these are based either on direct combustion or through fluidized bed com-bustion. Both these routes are beset with technical pro-blems because of the chemical composition of rice husk and its combustion characteristics2.

*For correspondence. (e-mail: anshu@cmu.edu)

RESEARCH COMMUNICATIONS

CURRENT SCIENCE, VOL. 87, NO. 7, 10 OCTOBER 2004 982

Rice husk is characterized by low bulk density and high ash content (18–22% by weight). The large amount of ash generated during combustion has to be continuously remo-ved for a smooth operation of the system. Frequently, the throat of downdraft gasifier gets clogged because of the sintering (caking) of the ash generated in the gasifier5. The rice husk gasifier virtually acts as a pyrolyzer since the residence time for the particles is often not long enough for char gasification reactions to proceed6. This results in poor carbon conversion efficiency7,8, sometimes as low as 55%. It has also been reported that the main reason for a poor utilization of carbon is due to difficulties in access-ing all the carbon present in the material. Some of it, trapped as chemical compounds in the ash, is not easily gasified or burnt2. Silicon oxide forms the main component (90–97%) of the ash with trace amounts2,7 of CaO, MgO, K2O and Na2O. The melting point of SiO2 is 1410–1610°C, while that of K2O and Na2O is 350 and 1275°C respectively7. It has been suggested that at higher temperatures, the low-melt-ing oxides fuse with silica on the surface of the rice husk char and form glassy or amorphous phases, preventing the completion of reaction9. This places an upper limit on local temperature of the gasifier10. Similar behaviour was also observed in fluidized beds where the bed materials, sand or alumina react with rice husk particles forming soft agglomerates and preventing the completion of pyrolysis and combustion reactions. Most of these observations were from bulk studies on rice husk and pyrolyzed products from gasifiers, as it was not possible to conduct in situ investigations on the pyro-lysis of individual particles under controlled heating and atmospheric conditions. The experimental difficulties in carrying out such studies have now been overcome with the availability of confocal scanning laser microscope (CSLM) with appropriate environmental chambers. In the following sections, we describe the results of pyrolysis and combustion experiments carried out on individual rice husk particles under controlled conditions in an environ-mental chamber attached to the CSLM. The results from these experiments were supplemented by SEM observations of the rice husk structure before and after pyrolysis. CSLM combines the advantages of confocal optics with a He-Ne laser, thereby enabling observations of samples at high resolution and at elevated temperatures. The con-focal optics enables the detection of strong signals from objects on the focal plane with weak ones from those not on that plane. By scanning a surface at various focal depths, a 3D image is constructed that faithfully records uneven surfaces. The utilization of a laser results in high illumination intensity that overcomes the noise generated by the thermal radiation present in materials at elevated temperatures. The optical set-up of this system thus has an improved signal-to-noise ratio, better contrast between different phases and higher resolution. For instance, this

microscope is capable of resolving features down to frac-tions of micrometres under controlled thermal fields and atmospheres, and can be operated up to 1700°C. Although confocal microscopy has been used quite extensively in biological research11, it is only recently that the laser integrated microscope with high temperature capabilities has entered into materials research. The CSLM at Car-negie Mellon, the only one of its kind in the US, is shown in Figure 1. Pulverized and sieved rice husk samples with sizes in the range between 250 and 300 µm were chosen for the experiments. The particles were roughly cylindrical with large aspect ratios. A single particle was placed in a cru-cible inside the furnace of the CSLM and was heated at a rate of 50°C/min in an inert atmosphere of flowing argon gas. The heating was mainly by radiation and the convec-tion effects were negligible. Argon was purified by pass-ing through a Cu-Mg heating coil. As a result, the oxygen partial pressure in the gas was reduced to 10–12 atm. The real-time visual images of the particle were recorded in a monitor, and still frames were later selected to analyse the particle shape and size as a function of temperature and time. This procedure was repeated for 20 randomly chosen particles of different shapes. Figure 2 shows the images of the same particle at different temperatures: 155, 432, 710 and 810°C. The above micro-structural observations show that there is virtually no change in size or shape of the particle until ~ 200°C. This is followed by a rapid shrinkage between 200 and 400°C. Above 400°C, the particle shrinks at a slow rate and stops shrinking after ~ 800°C. Two distinct stages in the shrinkage of particles are clearly seen (Figure 3 a), when the effective radius of the particles is plotted as a function of temperature. Mansaray and Ghaly12 reported a similar two-stage weight loss in thermo gravimetric experiments for different heating rates and atmospheres. The first stage can be explained as due to a rapid thermal degradation due to volatiles escaping from the particle. This process is complete when the temperature rises to ~ 400°C. The second stage is because of char combustion due to oxygen (present in argon) and gases generated from

Figure 1. Confocal scanning laser microscope at Carnegie Mellon University.

RESEARCH COMMUNICATIONS

CURRENT SCIENCE, VOL. 87, NO. 7, 10 OCTOBER 2004 983

the first stage. This stage is sluggish and less pronounced in rice husk. An important feature of this shrinkage is that the size reduction along the transverse direction of the largely longitudinal husks was significantly larger than along the longitudinal axis. Circumscribing an ellipse around the particle and calculating the aspect ratio captures this effect.

In Figure 3 b, the aspect ratio (L/D) increased from an initial value of 2.19 to 3.23. Figure 4 a and b shows the experimental results obtai-ned for several particles of different initial shapes and sizes examined in the study, but with a heating rate of 200°C/min. The trends are similar to the ones described above. The two-stage mechanism for particle shrinkage is

Figure 2. CSLM images of a rice husk particle heated in flowing argon at various temperatures. Particle shrinkage is much more in the transverse direction than in the longitudinal direction.

Figure 3. a, Effective radius of particle shown in Figure 2. b, Aspect ratio of particle vs temperature.

RESEARCH COMMUNICATIONS

CURRENT SCIENCE, VOL. 87, NO. 7, 10 OCTOBER 2004 984

clearly seen. Also, in all these cases, the aspect ratio in-creased, though the relative magnitude varied, probably due to the errors inherent in approximating the shapes of the particles to ellipses. These results support the hypo-thesis that there is a preferential direction for particle shrin-kage and this may be due to the structure of the silica skeleton in rice husk particles. Further heating to 1500°C did not show any change in the particle shape or size. This suggests that the residue is structurally strong and stable. This phenomenon appears to be unique to rice husk and some other straws. How-ever, most other biomass fuels do not exhibit this behavi-our7. The particle finally softened and melted at tempe-ratures above 1525°C (Figure 5). To gain additional insight into the particle’s structure, we performed scanning electron microscopic (SEM) observa-tions of a rice husk particle undergoing thermal degradation. To provide samples for SEM experiments, rice husk particles were heated to temperatures between 350 and 850°C in a tube furnace under an inert atmosphere of argon. The temperatures were chosen on the basis of the results obtained from CLSM experiments.

Figure 6 a shows the SEM image of a virgin rice husk particle. The structure resembles that of a composite mate-rial with fibres regularly interspaced in the matrix. In this case, the fibre is silica and the matrix consists largely of

Figure 4. Effective radius (a), and aspect ratios (b), of different particles at a heating rate of 200°C/min in flowing argon with an oxygen partial pressure of 10–12 atm.

Figure 5. Softening and melting of rice husk sample.

Figure 6. a, SEM image of the surface of a virgin rice husk particle showing longitudinal fibers interspaced within matrix of lignin and cellulose. b, SEM image of the cross-section of a virgin rice husk particle. Dense with no visible porosity.

a

b

RESEARCH COMMUNICATIONS

CURRENT SCIENCE, VOL. 87, NO. 7, 10 OCTOBER 2004 985

cellulose, hemi cellulose and lignin. Figure 6 b shows the cross-section of a virgin rice husk. The transverse section is dense, with no evidence of pores. The longitudinally aligned silica fibres can be seen in Figure 6 b and the dia-meter of theses fibres is between 5 and 10 µm. The SEM image of a rice husk particle heated to 350°C (Figure 7 a) shows the presence of a large number of but-ton-like structures or bumps interspaced with small pores. These were not present initially on the virgin particles. The presence of bumps and pores can probably be explai-ned by the volatiles escaping from the surface as a result of rapid thermal degradation between 250 and 400°C. The bumps could be due to the obstruction of silica fibre to devolatilization. The craters are the regions from where the volatiles have escaped the particle. The transverse-section (Figure 7 b) clearly shows the presence of pores. Pyrolysis of the cellulosic material has selectively consu-med the matrix leaving the silica fibres behind. Similar exploratory experiments with sawdust particles exhibited complete combustion and little residue remained. Figure 8 a and b shows the longitudinal and transverse sections of the particles heated to 850°C. The surface of the particle shows a large number of pores and bumps. This is similar to Figure 7 a, except that the number of

pores is more and their size larger. In Figure 8 b, the pores appear as channels from where the cellulosic material was preferentially removed during pyrolysis and char com-bustion. The CSLM experiments combined with SEM observa-tions suggest that the structure of rice husk is similar to a composite material with silica tubes filled with cellulose material constituting the strong-phase and the matrix con-sisting of lignin. There were no pores in the structure before gasification. However, after exposure to higher tempera-tures, the silica cylinders remain unaffected by gasifi-cation, and there are a large number of pores that are the remnants of pyrolysis and char combustion, These pores appear to have been arranged almost geometrically, sug-gesting preferential consumption of the matrix material during gasification. Kaupp7 also reported on the inertness and rigidity of the silica skeleton after thermal degrada-tion. The results on the microstructure of hollow fibres as seen in the transverse section of rice husk after exposure to high temperatures, and the structure of virgin rice husk with elongated silica fibres have not been reported ear-lier. These images also confirm the earlier observations of other researchers about the inertness of silica. It appears that the role of silica is more than just a geo-metric shield to the combustible material in the sample. If

Figure 7. a, Surface of a particle heated to 350°C showing large number of button-like structures or bumps interspaced with small pores. b, Cross-section of particle heated to 350°C clearly showing pores created as a result of pyrolysis.

Figure 8. a, Surface of rice husk heated to 850°C showing bumps and pores. b, Cross-section of rice husk particles heated to 850°C.

RESEARCH COMMUNICATIONS

CURRENT SCIENCE, VOL. 87, NO. 7, 10 OCTOBER 2004 986

silica were geometrically shielding the carbon, fluidized bed combustion could have yielded in better carbon con-version since it improves the comminuting of particles in the bed, thus exposing fresh surfaces for gasification. It appears that silica forms molecular bonds with carbon, which are not easily broken at the gasification tempera-tures. Reactions leading to the formation of silicon car-bide from silica involve very high temperatures (above 2500°C) and gasification temperatures are not high enough for such reactions to take place. In view of these results, we suggest that full carbon conversion in rice husk may not be achievable. However, this does not diminish the importance of rice husk as a fuel source because of its widespread and large avail-ability at affordable prices.

1. Jenkins, B. M., Physical Properties of Biomass, Gordon and

Breach, New York, 1989. 2. Natarajan, E., Nordin, A. and Rao, A. N., Overview of combustion

and gasification of rice husk in fluidized bed reactors. Biomass Bioenergy, 1998, 14, 533.

3. Department of Agriculture and Cooperation, Ministry of Agricul-ture, Government of India, New Delhi, 2000; http://agricoop.nic.in/.

4. Ministry of Non Conventional Energy Sources, Government of India, New Delhi, 2002; http://mnes.nic.in/.

5. Mukunda, H. S., Pers commun, Indian Institute of Science, Banga-lore, 2002.

6. Mukunda, H. S., Shrinivasa, U., Paul, P. J., Dasappa, S. and Rajan, N. K. S., Indian Institute of Science, Bangalore, 2002; http://fmb.no/ arnt/elspa/stand-alone.html

7. Kaupp, A., Gasification of Rice Hulls: Theory and Practice, Friedrick Vieweg & Son, 1984.

8. Flanigan, V. J., Xu, B. Y. and Huang, E., Tenth Annual Energy Sources Technology Conference and Exhibition, Dallas, TX, USA, 1987.

9. Bingyan, X. and Zongan, L., Advances in Solar Energy Techno-logy Biennial Congress, Hamburg, Germany, 1987.

10. Xu, B. Y., Huang, W. C., Flanigan, V. J. and Sitton, O. C., Sym-posium on Energy from Biomass and Waste IX, Chicago, IL, USA, 1985.

11. Sheppard, C. J. R., Confocal Laser Scanning Microscopy, New York, 1997.

12. Mansaray, K. G. and Ghaly, A. E., Agglomeration characteristics of alumina sand rice husk ash mixtures at elevated temperatures. Energy Sources J., 1997, 19, 989.

ACKNOWLEDGEMENTS. This work was funded by an United Nations Foundation Project. Defence Metallurgical Research Laboratory, Hyderabad provided SEM facility. We thank Prof. H. S. Mukunda, Indian Institute of Science, Bangalore for discussions and suggestions. Received 2 January 2004; revised accepted 8 June 2004

Accumulation of the periplasmic protein alkaline phosphatase in cell cytosol induces heat shock response in E. coli

Sujan Chaudhuri, Bimal Jana, Suchitra Sarkar and Tarakdas Basu* Department of Biochemistry and Biophysics, University of Kalyani, Kalyani 741 235, India

When the cells of Escherichia coli mutant MPh1061 (the mutation is at the N-terminal signal sequence part of the alkaline phosphatase [AP] gene) were grown in phosphate-less medium, accumulation of the AP pre-cursor (AP with its signal sequence) in the cell cytosol led to induction of cellular heat shock response. The autoradiograph of the SDS-polyacrylamide gel elec-trophoretic pattern of the proteins extracted from the 35S-methionine-labelled cells showed enhanced synthe-sis of different heat shock proteins. This was further confirmed by the immuno-precipitation study using the antibodies of the heat-shock chaperones, GroEL and GroES. Our experimental results suggest that the cytoplasmic accumulation of inactive AP behaved as abnormal protein to the cells, which triggered the induction of heat shock response.

IN Escherichia coli, all periplasmic and outer membrane proteins are synthesized in the cytosol initially in precur-sor form with an amino-terminal signal sequence of 18–26 amino acids1. The signal sequence binds to the trans-location machinery on the inner membrane and guides the precursor protein into the membrane for transloca-tion2. Moreover, interaction with cellular chaperones main-tains these precursor proteins in a translocation-compe-tent unfolded conformation until they are transported through the inner cytoplasmic membrane by the translo-cation machinery3. After the precursor proteins enter the periplasm, the signal sequence is ultimately removed by signal peptidase, an integral membrane protein with the active site facing the periplasm4. From analyses of the mutants, it is observed that a functional signal sequence is required for export; for proteins whose signal sequ-ences have been mutationally altered, their precursors accumulate in the cytoplasm instead of being exported out across the inner membrane5–7. In this study we have examined the expression of alkaline phosphatase (AP) in E. coli MPh1061 strain carrying a mutationally altered signal sequence (Leu-14 → Arg) to determine how E. coli cells respond to the storage of inactive AP precursor in cell cytoplasm by inducing cellular stress response. The bacterial strain E. coli MPh42 and its AP signal sequence mutant5 MPh1061, used in this study, were

*For correspondence. (e-mail: tbasu@cal3.vsnl.net.in)