Purchasing an electron microscope? – Considerations and scientific strategies to help in the decision making process

article
Gethin Owen
Resolve Microscopy Consulting Inc, Vancouver, British Columbia, Canada.
 
Biography
Gethin Owen has worked with electron microscopes for more than 20 years since completing a Biological Electron Microscopy MSc program in the UK. During this time he studied as a postgraduate in new imaging facilities both in Europe and North America, and experienced first hand the initial so called ‘settling in’ problems encountered with new electron microscopes after purchase and the steps needed to resolve these issues. In 2016 he founded Resolve Microscopy Consulting Inc., which offers a unique unbiased service to search for the ideal electron microscope based on the clients imaging needs and facility design with the ultimate goal of maximizing machine productivity and longevity.
 
Abstract
Electron microscopes are highly specialized tools, and are regarded to be a long term investment. With so many microscope models available, choosing a machine that will have the best sample specific imaging capability whilst working optimally in the facility environment, is no simple task.  This review attempts to summarize the necessary considerations before buying an electron microscope. Considrations include imaging performance, machine environmental condition require-ments, hardware and software, after sales support and servicing.  Following these considerations should ensure that the new machine provides problem free,optimal performance, once installed, for many years.
 
Corresponding author
Gethin Owen BSc (Hons), MSc, PhD
Resolve Microscopy Consulting Inc, Vancouver, British Columbia, Canada.
web: resolvemicroscopy.com
 
Introduction
 
Electron microscopes are highly specialized tools, and when purchasing such a machine it is generally regarded to be a long term investment with an approximate expected life-span of a minimum of 10 years. With so many microscope models available these days, as a buyer, choosing machines, is no simple task.
 
Many sales practices provide incentives for the buyer to purchase a machine which sometimes in hindsight is not appropriate for the needs of the microscope users and the over-all design of the facility. In many cases, the consequences of this situation are machine problems as a result of the lack of user expertise and sub-optimal environmental conditions. Both can initiate a snow-balling effect, resulting in the machine being down for servicing rather than up and running for imaging. Therefore, when buying a new machine, there should be a planned approach in testing the suitability of machines for the designated facility minimizing the chance of issues occurring.
 
In light of these above issues that I have personally encountered when in charge of purchasing an EM, I will attempt to provide a brief overview of the considerations that should be considered important before buying an EM.
 
To simplify the overwhelming amount of information I have divided the overview into three sections, which include microscope type, electron source and practical considerations and are arranged in order of importance respectively. However, it should be noted that all three sections are interlinked and will effectively be very relevant in the eventual performance of the microscope and consequently in the ultimate decision of which microscope and model to buy.
 
Microscope type 
 
Having personally had the opportunity to work in three new imaging facilities during my 20 years in the field of EM, I have realized that the sales person can have a huge influence in convincing the buyer what they need. 
 
The basic considerations that often influence the decision making process when purchasing a microscope are summarized in Figure 1. Nowadays, with more microscope choices available, electron microscope sales has become quite competitive between different manufacturers, therefore it is very important to carry out some research beforehand on all aspects involved in owning an EM so that the buyer is aware of the facts before purchasing.
 
Figure 1. A summary of the basic considerations that often influence the decision making process when purchasing an electron microscope
 
A more detailed model of the considerations that should be recommended when purchasing an EM can be seen in Figure 2. 
 
Figure 2. A detailed model of the considerations that are recommended when purchasing an EM
 
The first step in the process of buying an EM is to decide what type of imaging will be carried out in the facility. To help decide this, the types of samples that are going to be analyzed should be considered and what information will be collected i.e., sample surface information or internal features.
 
Currently, there are three options of microscopes that are commercially available, the Transmission Electron Microscope (TEM) and Scanning Transmission Electron microscope (STEM) for imaging and analysis of features from electron transparent or thin samples, the Scanning Electron Microscope (SEM) for imaging and analyzing the sample surface of bulk samples or the Dual Beam (SEM/Focused Ion Beam) system, that has the attributes of a multi-tool, capable of surface imaging with the SEM and creating cross-sections for viewing the internal features of the sample. 
 
Below, is a summary of generally available electron microscope models with consideration for the sample preparation techniques for each machine as well as the required supporting instrumentation for sample preparation and sample analysis.
 
Transmission Electron Microscope 
 
Transmission electron microscopes are designed to image samples in transmission therefore the samples have to be electron transparent and thin.  Two types of gun sources are available i.e. thermionic (which constitute, tungsten and lanthanum hexaboride sources) and field emission sources (either cold field emitter or Schottky) with accelerating voltages ranging from 120-300KeV. Specialized sample preparation techniques and hardware will be necessary to create electron transparent thin samples. The types of samples that can be imaged include ultrathin sections/cryosections, lamellae (created beforehand by Focused Ion Beam) and single particles (negatively stained, or vitreous ice-embedded cryo samples).
 
TEM’s can image hydrated (cryo) and dehydrated samples as well as having the capability to analyze samples analytically. Conventional imaging with such a machine, includes electron diffraction, high resolution TEM, low dose imaging, tomography, single particle analysis, negative staining, in situ imaging, and scanning transmission electron microscopy (in both bright field and dark field modes).  Numerous analytical analysis are also possible and include, energy dispersive x-ray microanalysis (EDX), electron energy loss spectroscopy (EELS), zero loss imaging, energy filtering TEM, cathodoluminescence as well as data acquisition automation by imaging software and robotic sample loading. The most recent model addition to the TEM family is an environmental TEM with the capability of in situ analysis in a gas controlled environment. 
 
Scanning Electron Microscope
 
Scanning electron microscopes are designed specifically to image the surface of bulk samples with the capabilities for imaging hydrated (cryo) and dehydrated samples. Thermionic, and field emission sources, identical to those found in TEM, are available. Environmental SEM (abbreviated ESEM) and or variable pressure (VP) systems have been recently added to the types of SEM models available so that minimal sample preparation is needed before imaging. 
 
Surface imaging is the ultimate goal of SEM, therefore it is necessary to be able to control the electron specimen interaction volume so that only the surface information is obtained. When testing SEM’s, it is essential to test the imaging capability of the machine at low accelerating voltages.
 
By convention it is claimed that anything less than 5keV is regarded as low voltage imaging but ideally around 1keV or lower should be expected. Surface analysis of the sample using secondary or backscattered electrons are generally the main surface imaging method but they have analytical capabilities with energy dispersive x-ray microanalysis (EDX) for elemental composition being the most common whilst electron backscattered diffraction (EBSD) for microstructure/crystallographic characterization and cathodoluminescent becoming more popular in recent years.
 
SEMs have more varied uses. Some SEM systems can be setup to image thin samples in scanning transmission electron microscopy (STEM) mode using a specially developed STEM detector and others are adapted to collect large areas of surface/elemental information through acquisition automation combined with image stitching software. 
 
Conventional SEM requires that the samples be dry and electrically conductive therefore sample preparation equipment will be needed to dry the sample and sputter coat non-electrically conductive samples with a thin layer of metal.
 
Nowadays, there are machines available using variable pressure (VP) with specialized electron detectors creating imaging conditions that allow biological samples to be examined with minimal preparation. Basic VP machines allows the operator to control the specimen chamber vacuum level but not to the point of imaging fully hydrated samples. Only the ‘environmental’ (ESEM) has this capability with the option of imaging specimens that are wet, uncoated or both by allowing for a gaseous environment in the specimens chamber. For high resolution imaging of hydrated biological samples in their native state, cryo attachments can be added to the machine, allowing hydrated samples to be imaged without drying artifacts. The cryo attachment will consist of a cryo chamber for freezing, a metal coating device to apply a conductive metal coating and a cold stage to keep the sample in the vitreous state during the imaging process.
 
See Table 1 for a concise summary of the different electron microscopes that are currently available commercially and the state of the art.
 
Table 1. A summary of the different electron microscope types that are currently available commercially. They are generally divided into 3 categories, transmission and scanning electron microscopes and dual beam systems. Modern transmission electron microscopes and scanning electron microscopes (SEM) are equipped to have the capability of imaging in different modes, and have diverse signal detection and analytical capabilities.  Dual beam systems have the capability of site specific sample preparation with the focused ion beam while viewing the process using the SEM. The state of the art nowadays is constantly changing with the goal of extending the imaging capabilities and resolution of the machines whilst increasing the throughput of data collection.
 
Scanning Electron Microscope/Focused Ion Beam
 
Dual Beams are comprised of an SEM and Focused Ion Beam (FIB). SEM’s on dual beams are identical to those found on regular SEM’s. Most manufacturers use field emitter Schottky electron sources however, cold cathode emission sources are available. 
 
Many ion beam sources are currently available and they differ in the element used to create the focused ion beam which ultimately determines the sputter volume capabilities which translates to the speed of milling and polishing resolution.
 
In most research settings, there are four sources, including gallium, xenon, neon and helium however, other specialist sources are available (Si, Au, Ge). Dual beams are used primarily for site specific processing which include TEM lamella prep (sample nano-manipulator), creating cross-sections, polishing, and patterning areas of interest but can be used for imaging using secondary ions. Gas injection systems are available for enhanced etch or deposition including proprietary software to monitor processing and endpoint.
 
Some manufacturers are developing multi beam FIB’s to gain the advantage of combining high precision beams with those capable of removing larger volumes increasing speed and throughput in the process.
 
Electron source
 
Once the type of imaging has been decided the next step in the decision process is to determine the image resolution requirements. Two emitter types are available, these being thermionic or field emitter, and they essentially determine the system resolution and therefore is reflected in the price and added complexity of the machine.
 
A thermionic gun has a large beam cross over diameter and therefore ideal for imaging large specimen areas, has consistent current density and stable emission, which is ideal for analysis techniques such as EDX. A field emission gun is used to produce an electron beam that is smaller in diameter, more coherent and with up to three orders of magnitude greater current density or brightness than can be achieved with conventional thermionic emitters. The result is significantly improved signal-to-noise ratio and spatial resolution, and greatly increased emitter life and reliability compared with thermionic devices.
 
The field emitter gun is the high resolution option and is the most expensive option because of its high performance. However, the significant costs are not only in the machine price but after purchase costs. Being a high resolution system, environmental conditions influence the machine performance and therefore machine location and the necessary infrastructure becomes very important. In addition, such a machine generally requires peripheral systems such as uninterruptible power supply, water chillers, and scroll pumps, housed separately from the machine, and depending on the sample, there may well be a need for additional lab space with high performance sample preparation equipment. Moreover, daily consumables such as high purity gases, essential for high performance scroll pump gas ballast and chamber venting, should be considered as well as the dedicated time for daily preventative maintenance and beam/lens alignment procedures. 
 
Practical considerations
 
Once the machine type and detectors for sample analysis have been decided there are three main areas to research before narrowing down the choice of the manufacturer and model. These three areas are: machine performance on paper, location suitability and sample dependent machine performance. 
 
In view of these criteria, I will attempt to highlight the key practical issues that should be evaluated when choosing an electron microscope in the next section.
 
1 Machine build – most microscopes will be operated in spaces with considerations for optimal imaging conditions but in some cases machines can be operated in quite harsh conditions. In this case the first consideration will be the robustness of the outer skeleton and its ability to withstand heavy use. Interestingly, when inspecting machines from different manufacturers some designs are more robust than others. Also, as the state of the art boundaries are constantly changing, the age of the technology becomes a very important factor.
 
The age of the technology for the electron source, lenses, and vacuum system should be investigated to ensure that the latest technology is being purchased to ensure product longevity. Other factors that are considered important are stage load capacity, sample size, stage tilt, precision and stability in x,y,z and stage eucentricity in particular for dual beam systems. This will of course be dependent of the type of samples and imaging required.
 
Other practical characteristics include simplicity of sample loading, speed of sample loading, amount of user involvement for failsafe unsupervised usage, minimum and maximum working distance in high resolution mode to establish practical sample size, high throughput methods (chamber pump down/vacuum recovery speed, number of samples per holder) and available sample navigation devices.
 
2 Machine environment – the performance of any EM is influenced by magnetic fields, vibrations, barometric pressure changes, room temperature variations and water chiller temperature variations. To maximize performance it is necessary to exceed the manufacturers’ requirements to minimize adverse ambient conditions. 
 
The building location and room in which the instrument is housed are two factors that are really the main determinants of machine performance.  Vibration from the surrounding floor and air buffeting the machine within the room are major factors in diminishing machine performance. Stray magnetic fields can also influence performance. The most practical way to get over these problem are to isolate the machine from them. For more in-depth information see 'Suggested reading'. The most common methods are summarized below:
 
a Vibration – vibration generally comes from the floor (traffic or from microscope ancillary equipment) or air (acoustic or vibrations). It is important to make sure the machine is isolated from vibration emanating from the floor by mounting the machine on a heavy concrete pad. Nowadays, most machines, particularly with field emission sources, are also required to have a separate mechanical room to hold microscope ancillary equipment such as pumps and compressors, store gas cylinders and water chillers that maintain a constant temperature for the electromagnetic lenses.   
 
b Magnetic field – electromagnetic interference must be minimized as these can cause aberrations in high resolution TEM, scanning distortions in STEM images and loss of resolution in electron energy loss spectroscopy. All electrical conduits, fire sprinkler mains, duct work, chiller water ducts, hot water lines and drains should be routed around the room. There should be no electrical or water lines directly above the instrument.
 
c Temperature and humidity – there are other factors that are overlooked that will eventually limit the machine performance on a day to day basis other than vibration and stray magnetic fields. In my experience heat is one of the main enemies of the electron microscope.
 
Keeping the machine temperature constant is extremely important and modern machines generate a tremendous amount of heat from associated servers and computers that are used to control the machine. Variations in air temperature causes specimen drift in the microscope electronics, changes in mechanical tolerances of components e.g. microscope lenses, correctors and scan coils, ultimately decreasing the imaging capabilities of the instrument. Unfortunately, dissipating this heat is difficult in rooms that conform to the minimum guidelines often created by manufacturers. In this respect it is recommended that rooms should be larger than the minimum recommended requirements. 
 
Moreover, in-house air conditioning systems are insufficient in maintaining the temperature and relative humidity for sensitive EM’s. More often than not they cause more unnecessary issues such as when the air flow is increased to cool down the room creating air currents that cause acoustic vibrations. Optimally, air handling should be designed to deliver just enough cold air to balance the heat production of the source in the microscope room (remember to consider the operator heat production) to limit cycling between temperature extremes.
 
Such in-house systems rarely work therefore a separate air handler that serves the room with dedicated supply and return duct work should be used. Therefore, an investment should be made to purchase individual air handing units whilst minimizing air currents in the room (duct sock) or alternatively add thermal mass to the room with radiant cooling. In the long run this is a worthwhile investment to safeguard the machine performance over its lifetime.
 
d Air pressure fluctuations – air pressure changes can cause machine instabilities during high resolution imaging. Such pressures occur due to the air conditioning system over pressuring the building, as well as the enclosed rooms within, compared to ambient pressure. Consequently, opening doors can send a pressure wave caused by the pressure differential between areas, ultimately generating shifts in the image during acquisition. These issues generally arise with high resolution machines where image acquisition is by electron beam scanning e.g. STEM and quite possible these days with FEG-SEM’s. 
 
3 Microscope Company an electron microscope is a long -term investment therefore the manufacturer needs to be reputable with a history of producing machines for some time. This should guarantee the availability of parts in the long term if needed. In the current global business world it is pertinent to also determine whether the microscope manufacturer is based in your country or is the manufacturer selling its machines through a distributor?
 
Some questions that are then important to ask is if the machine is sold by distributors are: who will service the machine; who does the distributor sales department work with when it comes to adding periphery devices?
 
4 Servicing/ maintenance
a Service contract – when purchasing an expensive complex machine such as an electron microscope it is necessary to have a service contract which should be negotiated during the purchasing process. Service contract cost for thermionic systems are low whereas field emission systems are much higher and increase with the complexity of the instrument. Before agreeing on a service contract it is important to make enquiries whether the maintenance will be done by the manufacturer factory trained technicians or a third party contractor. Background checks on the expertise of the service engineer should be carried out, base location, and the servicing area that the engineer covers.
 
One other important thing to consider is whether the response service time on the contract is practically feasible i.e. can the engineer be physically present to fix the issue within the guaranteed response time-line on the service contract. In cases where the service coverage area by the engineer is enormous, the travel time needed would make it virtually impossible to achieve the guaranteed service time. Keep this in mind when reading the service contract coverage. 
 
b Online connectivity – these days most machine issues can be resolved through remote access and diagnostics. Having the appropriate room connections, internet service and network security is a prerequisite for this sup-port and should be organized with information technology personnel be-forehand. 
 
c Additional maintenance costs – preventative maintenance is essential for the machine to be functional but unless specified, service contracts do not cover consumables and these costs should be clarified before planning your annual budget. 
 
Other surprise costs can also come up when replacing sources e.g. when replacing a focused ion beam tip on some machines it is recommended to replace the suppressor and extractor every other tip replacement. These add a significant cost to a tip replacement e.g. a tip replacement cost can almost triple in cost!
 
5 Microscope computer hardware and software – a significant portion of the machine operation depends on software, therefore, the user interface should be intuitive to experienced microscopists but also simple enough for novice users.  User interfaces vary between manufacturers and are generally developed by programmers, not microscopists and offer many ways to carry out the same step (which are practical and sometimes impractical). 
 
This is very much a factor that needs careful consideration if the machine is to be used in a multi-user/teaching facility.  Moreover, very often it is taken for granted that both the computer hardware and operating system can simply be upgraded without adversely affecting the microscope control software. This should not be taken for granted!
 
Constant improvements in operating systems necessitate more frequent upgrades and the purchasing of more powerful computer hardware. Keeping up with the new releases in operating systems can be an added future cost due to compatibility issues with the microscope control software. Therefore, it is always advisable to be prepared by researching the future cost of upgrading the microscope control software. 
 
6 Data analysis software – post analysis of data following acquisition on the instrument is often needed in EM labs. It is worth considering purchasing the software as a bundle with the instrument for significant discounts, as the cost may increase considerably thereafter.
 
7 Sample preparation equipment – the quality of the data collected with electron microscopes are dependent on the preparation method used. Sample preparation for electrically conductive materials may require minimal preparation however, for hydrated non-conductive samples consideration should be given to use preservation techniques that are capable of preservation finer than the resolution of the microscope. The capabilities of the preparation machines should be considered in the purchasing process.
 
8 Running costs – yearly renewal of the service contract, consumables, down time/loss of use for servicing, weekly alignments, cleaning, testing are factors that should be considered in the choice of machine. This is especially true when the uptime of the machine is important e.g. industry quality control and fee for service based imaging facilities.
 
9 Specialist technician – electron microscopes will vary in complexity and therefore the level of upkeep will differ. Most modern basic thermionic machines are very stable and require little or no maintenance. For the more complex field emission systems they will require constant monitoring on a daily basis and will likely require a technician to keep the machine in optimum working order. The cost of additional personnel should be considered in the budget during the purchase process. 
 
10 Impartial testing methods for the microscope – very often electron microscopes are marketed by image performance using ideal samples. However, the microscope performance is well known to be sample dependent and this aspect can be used to test the performance of the machine under different conditions.
 
Machine performance by manufactures is often demonstrated using ideal samples. Gold coated on carbon samples is a sample widely used because it is a sample with high conductivity and has a good range of contrast levels. These demonstration samples do not show real-world performance.
 
When I perform a series of tests for clients, I generally choose samples that specifically could determine the performance of individual components of the machine as well as imaging. Generally, around 6-8 samples should suffice and these can be prepared beforehand and sent as standardized samples to different manufacturers. In the past the manufacturers would pay for a prospective buyer to travel to a facility or their headquarters and use the machine for testing. In this day and age this is not essential and the machine can be tested live online, either by instructing a technician operating the machine through a live feed or through virtual electronic control of the machine. Once data for each sample has been acquired they can be analyzed and compared. This factor can help in the impartial testing process.
 
11 References – finally the last form of testing is to reach out to local and national users that have purchased machines from manufacturers and see what kind of support they have had working with the manufacturer after purchase, as well as machine performance and reliability. More than likely a list of users can be provided by the sales personnel.
 
Summary and conclusion 
 
Purchasing a new electron microscope is quite a responsibility these days. So much stress on maximizing investment over a limited time span as the technology is constantly changing.
 
Conventional practices of purchasing such a machine following past methods are no longer applicable and the decision should be based by thorough testing in a logic and scientific way.
 
Deciding on a particular machine requires a lot of research on behalf of the buyer before approaching sales as well as developing a standardized impartial testing method for narrowing down the machine choices. 
 
Choices  should be based not only on imaging performance but machine environmental condition requirements, hardware and software associated with the machine as well as after sales support and servicing to make sure that the machine is functioning properly once installed. A general guideline, organized as a flow diagram, showing the steps in the process of purchasing an electron microscope is provided in Figure 3. Following these guidelines and consideration should direct new buyers in the right direction in purchasing a machine that provides problem-free optimal performance for the lifetime of the machine.
 
Figure 3. A general guideline, organized as a flow diagram, showing the recommended steps in the process of purchasing an electron microscope
 
Suggested Reading
  • Martinez-Tejada HV (2014) General considerations for design and construction of transmission electron microscopy laboratories. Acta Microscopica 23:56-69.
  • Muller DE, Kirkland E, Thomas MG, Grazul JL, Fiting L, Weyland M (2006) Room design for high performance electron microscopy. Ultramicroscopy 106:1033-1040.
  • O’Keefe MA, Turner JH, Musante JA, Hetherington CJD, Cullis AG, Carragher B, Jenkins R, Milgrim J, Milligan RA, Potter CS, Allard LF, Blom DA, Degenhard L, Sides WH (2004) Laboratory design for high-performance electron microscopy. Microscopy Today 12:8-14.
  • Muller DA, Grazul J (2001) Optimizing the environment for sub 0.2nm scanning transmission microscopy laboratories. J Electron Microsc (Tokyo) 50:219-226.
Website developed by S8080 Digital Media