ACTA BIOLOGICA SLOVENICA LJUBLJANA 2003 Vol. 46, Št. 2 : 11 - 15 Sprejeto (accepted): 2003-10-30 Investigation of Plant Surfaces with Environmental Scanning Electron Microscopy (ESEM®) - A Comparison with Conventional SEM Dagmar KOLB, Edith STABENTHEINER* Institute of Plant Physiology, Karl - Franzens University Graz, SchubertstraBe 51, A-8010 Graz, Austria Introduction *corresponding author (e-mail: edith.stabentheiner@uni-graz.at) Abstract. Environmental scanning electron microscopy (ESEM) enables the investigation of untreated and watercontaining material without preparation with the benefit of SEM (depth of focus and three dimensional imaging of surfaces with a high resolution). Conventional SEM (CSEM) usually requires tirne consuming fixation, drying and coating of samples. Their surlace structures may be altered by this procedure. For comparison a large number of plant samples was observed with both methods. Using CSEM, secretion products or mucilagineous coatings may be removed and dynamic processes cannot be observed. However, the samples can be investigated several times. In contrast, ESEM allows the observation of watercontaining, native surfaces and this method is the only possibility to watch dynamic processes in the SEM. However, using ESEM the plant material is very sensitive to beam damages because of the lack of the protecting metall layer - necessary for non-conducting surlaces in CSEM and dehydration cannot be prevented completely. In summary, ESEM will not compete with CSEM but it will establish oneself as a valuable and essential supplement in studying plant surlaces. Key words: environmental scanning electron microscopy (ESEM), conventional scanning electron m.icroscopy (CSEM), plant surlaces Scanning electron microscopy (SEM) became an indispensable tool in studying plant surlaces. For conventional SEM (CSEM) biological samples usually have to be fixed, dehydrated and coated (ROBINSON & al. 1987, DYKSTRA 1992). The environmental SEM (ESEM) allows the observation of many types of specimens without subjecting them to conventional preparation techniques (DANILATOS 1993). This is possible because of a pressure limiting aperture with high vacuum maintained in the beam-generating and- focusing part of the column (- 104 Pa in the gun area), while low vacuum (up to 102- 103 Pa) is tolerated in the specimen chamber (BozzoLA & RussEL 1992, DYKSTRA 1992, DANILATOS 1993). The secondary electrons em.itted from the sample collide with water molecules in the chamber so as to produce additional electrons and positive ions. The positive ions are attracted to the sample surlace and eliminate charging artefacts. This ionization process results in a proportional cascade amplification 12 Acta Biologica Slovenica, 46 (2), 2003 of the original SE signal which is detected by a special gaseous secondary electron detector (DANILATOS 1993, TAr & TANG 2001). Asa consequence, unfixed and uncoated samples - even those containing considerable amount of water - can be investigated by SEM. To compare CSEM and ESEM various plant samples were investigated on the one hand after fixation, drying and coating and on the other hand without any preparation at differing ESEM conditions. Material and Methods A large number of different plant samples was investigated using the following microscope conditions (KOLB 2002): CSEM: conventional SEM with high vacuum (- 10·4 Pa) in the chamber; sample preparation: chemical fixation (e.g., glutaraldehyde), dehydration, critical point drying with CO 2 as drying agent, sputtercoating with gold (ROBINSON & al. 1987). ESEM: a) sample temperature 5°C (Peltier cooling stage); gaseous secondary electron detector; chamber pressure 133-930 Pa; relative humidity: up to 100 % (Table l); no sample preparation. b) samples at room temperature (without cooling); gaseous secondary electron detector; chamber pressure 133-670 Pa; relative humidity < 20 % (Table l); no sample preparation. c) large field gaseous secondary electron detector (pressure limiting aperture with wider diameter than a) and b); pressure in the charnber maximally 133 Pa; relative humidity < 1 O % (Table 1 ); no sample preparation. All samples (CSEM, ESEM) were mounted on aluminium stubs with double sided conductive tape and were investigated at different microscope conditions with a Philips XL30 ESEM using an acceleration voltage of 20 kV. Table 1: Values show chamber pressure in Pa, corresponding relative humidity (%) at sample temperatures (0 C) from 0° to room temperature (25°C). RH 100 % 80 % 60 % 40 % 20 % 10 % Temperature oo 612 480 360 239 120 67 50 865 692 519 346 173 93 100 1224 971 732 492 239 120 15° 1702 1357 1024 678 346 173 20° 2328 1862 1397 931 466 239 25° 3152 2527 1889 1264 625 319 Results and Discussion The environmental scanning electron microscope (ESEM) allows the observation of many types of specimens without subjecting them to conventional preparation techniques because it allows the introduction of a gaseous environment in the specimen chamber (DANILATOS 1993, TAr & TANG 2001) . To compare CSEM and ESEM a great many different plant samples were investigated on the one hand after fixation, drying and coating and on the other hand without any preparati on at differing ESEM conditions. All samples - without any restrictions - could be investigated with conventional SEM (CSEM) using chemical fixation followed by dehydration, drying and coating procedure (Figs. 1-3). Leaves and shoots of plants are easy to handle during preparation. Very small samples, e.g. unicellular algae (Fig. 1, Micrasterias sp.) were attached to cover slips with poly-L-lysine prior to the preparati on procedure (ROBINSON & al. 1987). Generally, no artefacts due to beam damage or due to insufficient D. Kolb, E. Stabentheiner: Investigation of Plant Surfaces with Environmental Scanning .. . 13 electrical (charging effects) and/or thermal conductivity occurred. Samples can be stored for a long tirne under appropriate conditions (dry and clean atmosphere) and can be investigated as often as necessary (CRANG 1988, DYKSTRA 1992). 14 Acta Biologica Slovenica, 46 (2), 2003 Figs. 1-3 CSEM samples. Fig. 1: Micrasterias sp. bar= 50 µm; Fig. 2: a stigma of Setcreasea purpurea with its stigmatic hairs bar = 200 µm; Fig.3: a preparation artefact of a calcareous crust surface of Saxifraga kolenitiana bar= 10 µm; Figs. 4--10 ESEM samples. Fig. 4: a pollen grain interaction with stigmatic hairs bar= 50 µrn; Fig. 5: a turgeszent trichome of Lycopersicon esculentum bar= 20 µm; Fig. 6: a native state of Micrasterias sp. bar= 50 µm ; Fig.7: a stigma of Setcreasea purpurea with mucus bar == 100 µm; Fig. 8: tbe native state of a calcereous crust of Saxifraga kolenitiana bar = 20 µm; Fig. 9: water droplets on the surface of Drosera rotundifolia bar== 100 µm; Fig.10: an example of surface debydration and cbarging effects of Melissa officinale bar== 20 µm. In contrast, ESEM allows tbe observation of biological samples in tbeir natura! state and witbout coating (Figs. 4--8; DANILATOS 1993, TAI & TANG 2001, Y AXLEY & al. 2001). Typically, the first steps in sample preparation for CSEM are fixation and dehydration (ROBINSON & al. 1987). However, these steps often result in a removal of surface coatings (CRANG 1988). In Fig. 7 tbe mucilaginous coating on a stigma of Setcreasea purpurea can be observed investigating fresh samples (ESEM), whereas this coating is removed after preparation for CSEM (Fig. 2). However, the stigmatic hairs are hidden on the fresh surface and can only be investigated in detail after sample preparation. So both methods complement one another. A well known artefact due to dehydration is sbrinkage of up to 40 % ofthe original volume (CRANG 1988). This can be clearly demonstrated bere wben comparing the sample with (Fig. 2, bar== 200 µm) and without (Fig. 7; bar== 100 µm) preparation. Only ESEM allows the investigation of tbe close interactions between pollen grains and stigmatic hairs of Hibiscus sp. (Fig. 4). Besides the removal ofcoatings surface deposits can be modified. In Fig. 8 the native state of tbe calcareous crust on leaves of Saxifraga kolenitiana can be observed wbile in Fig. 3 tbe structure of tbe crust was altered due to sample preparati on. Investigating biological samples containing a considerable amount of water a cooling stage belps to control the temperature of tbe specimen and thus the relative humidity, whicb is a strong function of the temperature (Table 1; DANILATOS 1993). Cooling enables tbe maintenance of a bigb relative bumidity on the sample surface (Table 1). As a consequence debydration is prevented and even delicate plant structures can be observed in tbeir native state for up to 60 minutes (Figs. 4--8). Different types of plant hairs can be easily observed without any preparation (Fig. 5). Even very sensitive algae as Micrasterias sp. (Fig. 6) can be investigated when sufficient water supply from a wet filter paper or agar is ensured. However, care bas to be taken that cbamber conditions are controlled in a way tbat bumidity on tbe sample surface is bigb enougb to stop dehydration but not too bigb to produce water droplets on tbe surface making it invisible (Fig. 9). The following conditions tumed out to be optimal for the investigation of wet samples: 5°C sample temperature and approximately 640 Pa vapour pressure. Tbese conditions are very sirnilar to those found by TAI & TANG 2001. The possibility to control relative bumidity enables tbe direct investigation of dynamic processes on plant surfaces, e.g. debydration and rebydration cycles as it was done investigating the swelling bebaviour of cellulose fibres (JENKINS & D0NALD 1997). However, if not undisturbed samples ( e.g., wbole leaves) but sliced samples are used dehydration can not be completely prevented. Shrinkage and cbarging are tbe consequence (Fig. 10). ESEM witbout cooling results in quite low relative bumidities on the sample surface (Table 1). Many plant structures, especially tbose equipped with thick celi walls and cuticles, can be investigated in tbis mode giving tbe same results as cooling tbe samples. Investigation tirne, bowever, is mucb shorter and debydration occurs much faster. In contrast to CSEM, wet samples can only be used once - for further investigations a new sample is necessary. Tbe samples are more easily damaged by beam current and accelerating voltage since no coating is present that ensures sufficient tbermal conductivity and stabilization of the surface (CRANG 1988). Anotber disadvantage of ESEM is a reduced field of view at lower magnifications D. Kolb, E. Stabentheiner: Investigation of Plant Surfaces with Environmental Scanning ... 15 due to the pressure limiting aperture (Fig. 9). This can be overcome using the large field gaseous secondary electron detector (FEI). However, due to the wider diameter of the aperture the maximal chamber pressure is restricted to 133 Pa and asa consequence the relative humidity on the sample surface is rather low (Table 1). So the same restrictions for fresh samples as stated above goes for this mode of ESEM. The mentioned restrictions are not applied to dry and stable samples (e.g., wood, insects). Conclusions ESEM represents a step forward in the instrumentation of electron microscopy and it allows access to areas of research not previously possible (DANILATos 1993). However, it will not compete with CSEM but it will establish oneself as a valuable and essential supplement in studying plant surfaces. Literature BozzoLA J. J. & L. D. RussEL 1992: Electron Microscopy. Jones and Bartlett Publishers, Boston- London. CRANG R.F.E. 1988: Artifacts in specimen preparation for scanning electron microscopy. In: Crang R. F. E. & K. L. Klomparens (eds.): Artifacts in Biological Electron Microscopy, Plenum Press, New York - London pp. 107-129. DANILATOS G. D. 1993: Introduction to the ESEM instrument. Microscopy research and Technique 25: 54-361. DYKSTRA M. J. 1992: Biological Electron Microscopy. Plenum Press, New York-London. KoLB D. 2002: Untersuchungen an frischem und fixiertem Pflanzenmaterial mit Hilfe des Rasterelektronenmikroskops, Diplomarbeit, Karl - Franzens - Universitat Graz. JENKINS L. M. & A. M. DoNALD 1997: Use ofthe Environmental Scanning Electron Microscope for the Observation of the Swelling Behaviour of Cellulosic Fibres, Scanning 19: 92-97. ROBINSON D. G., U. EHLERS, R. HERKEN, B. HERRMANN, F. MAYER &F.-W. ScH0RMANN 1987: Methodes of Preparation for Electron Microscopy, An Introduction for the Biomedical Sciences, Springer - Verlag Berlin Heidelberg New York. TAI S. S. W. & X. M. TANG 2001: Manipulating Biological Samples for Environmental Scanning Electron Microscopy Oberservation. Scanning 23: 267-272. YAXLEY J. L., W. JABLONSKI, J.B. Rmo. 2001: Leaf and flower development in pea (Pisum sativum L.): mutants cochleata and unifoliata. Annals of Botany 88: 225-234.