Atom Probe Tomography – a high throughput screening tool of 原子探针断层–高通量筛选工具

1、198Combinatorial Chemistry & High Throughput Screening, 2011, 14, 198-2051386-2073/11 $58.00+.00 2011 Bentham Science Publishers Ltd.Atom Probe Tomography A High Throughput Screening Tool for Atomic Scale ChemistryKrishna Rajan*Department of Materials Science & Engineering and Institute for

2、Combinatorial Discovery, Iowa State University, Ames, IA 50011, USAAbstract: The objective of this paper is to examine the challenges and opportunities in high throughput screening of atomic scale chemistry via a technique known as atom probe tomography. While there are numerous methods that exist i

3、n the field of throughput screening, even at the nanoscale, most of the effort is on screening properties, rather than chemistry and/or structure. In this overview, we discuss the role atom probe tomography can have as a high throughput screening tool of atomic scale chemistry. Particular emphasis o

4、n the study of organic/biological materials is given along with the needs and challenges to make atom probe tomography more widespread in the field of combinatorial chemistry.Keywords: Atom probe tomography, atomic scale chemistry, data, informatics, organic films.INTRODUCTION TO ATOM PROBE TOMOGRA-

5、PHY: HIGH THROUGHPUT DATA ACQUISITIONAtom Probe Tomography (APT) represents a revolutionary characterization tool for materials that combines atomic imaging with field ion microscopy (FIM) with a special time-of-flight (TOF) mass spectrometer to provide direct space three-dimensional, atomic-scale r

6、esolution images of materials with the chemical identities of all the detected atoms. It involves the controlled removal of atoms from a specimens surface by field evaporation and then sequentially imaging and analyzing them with a TOF mass spectrometer. Atoms are selected from a region on a specime

7、ns surface (the area may be as large as 200 nm x 200 nm) and are then spatially mapped (see Fig. 1). When combined with depth resolution of one inter-planar atomic layer for depth profiling, an APT provides the highest spatial resolution of any microanalysis technique. This capability provides a uni

8、que opportunity to study experimentally with atomic resolution, chemical clustering and 3-D distributions of atoms, and directly test and refine atomic and molecular based modeling studies. While APT has its origins in the field-ion microscope (FIM), originally developed by Erwin W. Mller in 1955 an

9、d the atom probe microscope dates back to ca. 1968, it is only fairly recently that highly sophisticated and reliable instruments have become commercially available. Improvements in data collection rates, field-of-view, detection sensitivity (at least one atomic part per million), and specimen prepa

10、ration have advanced the atom probe from a scientific curiosity to a state-of-the-art research instrument 1-10.When one considers the term “high throughput screening” in the context of combinatorial experiments, one usually envisages techniques that scan a two dimensional area that encompasses a lib

11、rary of materials chemistries. This includes a wide array of techniques ranging from X-ray *Address correspondence to this author at the Department of Materials Science & Engineering and Institute for Combinatorial Discovery, Iowa State University, Ames, IA 50011, USA; Tel: 515-294-2670; E-mail:

12、 diffraction to a variety of scanning probe methods. A fundamental challenge in all these methods, is that high throughput does not necessarily translate to high resolution from of a microstructural, crystal structure or electronic structure perspective (see Fig. 2). Despite the vas

13、t amount of work in the field of combinatorial materials science and chemistry, characterizing information with high resolution while acquiring data very fast is still a challenge.Fig. (1). The specimen (#1) is inserted into a cryogenically cooled, UHV analysis chamber. The analysis chamber is cryog

14、enically cooled to freeze out atomic motion. It is at ultrahigh vacuum (UHV) to allow individual atoms to be identified without interference from the environment. A positive voltage is applied to the specimen via a voltage/laser pulse. The positive voltage attracts electrons and results in the creat

15、ion of positive ions. These ions are repelled from the specimen and pulled toward a position-sensitive detector. The location of the atom in the specimen is determined from the ions hit position on the detector (#2). This configuration magnifies the specimen by a million times and in due course, ato

16、ms from the surface ionize, exposing another layer of atoms under them. This process of field ionization continues until the specimen has been fully analyzed, and provides a 3D image of the entire specimen (Courtesy: D. Seidman).In this paper, we want to discuss a characterization approach that prov

17、ides a new and exciting approach to high throughput screening of atomic scale chemistry; atom probe tomography. The term high throughput screening comes into Atom Probe TomographyCombinatorial Chemistry & High Throughput Screening, 2011, Vol. 14, No. 3 199play here in the context that it refers

18、to the nature and speed of detecting massive amounts data from individual atoms/molecular fragments. A typical analysis with a three-dimensional atom probe data can contain up to half a billion atoms if not higher and the data acquisition rate can be easily 100 million atoms a day. Also as is discus

19、sed below, that APT can be integrated well into the study of two dimensional chemical library arrays 11.A key challenge for atom probe tomography is to fabricate a sample into a needle shape with tip radius of 50 nm. For the study of organic materials, probably the most efficient way is to coat a sh

20、arp pre-fabricated needle 12. A microtip array is an ordered collection of tall (100 micron) Si posts formed on a Si coupon with standard microelectromechanical systems (MEMS) techniques. The array enables the atom probe to field evaporate atoms exclusively from a single needle-shaped specimen locat

21、ed among a field of similar needle-shaped specimens. These microtip arrays enable the rapid preparation of a large number of nearly identical atom probe samples by providing a consistent platform for mounting the sample section once it is extracted from the material of interest. This provides a usef

22、ul way of developing combinatorial arrays of atom probe specimen.In the following discussion, we review the limited amount of work in the open literature that addresses the application of atom probe tomography to the study of organic/biological materials, especially from the perspective of high thro

23、ughput analysis. This is followed by a summary of the data analysis and specimen preparation challenges associated with biological systems for this kind of study.ATOM PROBE TOMOGRAPHY AS HIGH THROUGHPUT SCREENING TOOLNormally atom probe tomography is not considered as a screening tool for combinator

24、ial libraries. However, when one considers the process of high throughput screening of combinatorial libraries, one critical issue that is often overlooked, especially in terms of chemical/structural characterization of libraries, is the trade-off between speed and spatial resolution/sensitivity of

25、the analysis. This problem becomes even more acute as the spatial density of the arrays decreases. With the capacity of using nanoscale lithography techniques as shown in Fig. (3), one can produce a very high density of “nanoarrays” instead of the usual microarrays. Each tip (with 50 nm radius) can

26、serve either as a substrate for generating gradient thin film structures with each needle associated with a specific chemistry or this array can be extracted, using focused ion beam techniques from a bulk material with known chemical and/or microstructural gradients. Normally in 2 dimensional combin

27、atorial libraries, a scanning probe techniques often provides the key means of screening the library. With the atom probe, we have an in a sense a “reverse” geometry where the specimen to be screened needs to be in the form of a tip and we need a way to characterize at the tip of each specimen with

28、extremely high resolution and sensitivity. As show in Fig. (2), atom probe tomography provides us the highest level of both spatial resolution and chemical sensitivity of any technique to analyses materials chemistry. Hence library geometries that lend themselves well to the Fig. (2). A comparative

29、chart showing the value of atom probe to achieve both chemical resolution and chemical sensitivity relative to other high throughput atomic scale chemical characterization techniques.200 Combinatorial Chemistry & High Throughput Screening, 2011, Vol. 14, No. 3Krishna Rajanuse of atom probe tomog

30、raphy as a high throughput screening tool for combinatorial arrays include:Fig. (3). A silicon microtip array of atom probe tips for use as a platform for depositing combinatorial arrays of film overlays on thin tips 12.Thin film gradients across nanoarrays substrate tips;Developing diffusion based

31、phase libraries on each tip to add a third dimension to our ability to create compound libraries. In this case, we use the atom probe to also interrogate chemistries in the vertical direction along the axis of each specimen tip. An example of this is shown in Fig. (4) below for the detection of comp

32、ound phase libraries based on interdiffusion between thin film overlayers and the substrate; andThe use of nanolithography techniques to extract nanoarrays of atom probe tomography suitable tips from a system with a known combinatorial array or gradient of chemisty and or microstructure.HIGH THROUGH

33、PUT MOLECULAR SCREENING OF ORGANIC FILMSAs noted in the recent National Academy report on “A New Biology for the 21st Centurythe advancement of imaging techniques of biological materials based on cryogenic preparation techniques (e.g. cryogenic electron tomography) offers opportunities for “unpreced

34、ented insight into the molecular organization of cellular landscapes” 13. Atom probe tomography (APT) that can provide nearly an order of magnitude image resolution improvement compared to high resolution TEM, and yet capture chemical information competitive with other structural analysis tools such

35、 as X-ray diffraction. It can provide biologists with a powerful new technique for characterizing molecular structure and for understanding in detail the mechanisms of chemical processes in biological complex systems that present microscopy methods cannot provide. While mass spectral analysis of org

36、anic films using APT has been reported in the open literature, biological imaging has not. We shall in the following discussion review some of the work on organic films and conclude with a discussion of the challenges and opportunities for high throughput screening of atomic scale information of bio

37、logical systems.Fig. (4). An atom probe based profile of a combinatorial thin film library of multilayer films from a nanoarray of atom probe tips as shown in Fig. (2); detecting phase formation of Cu3Si through thin film diffusion couples. Such unprecedented level of spatial resolution and chemical

38、 sensitivity will not be possible by other techniques used for chemical screening of combinatorial libraries (from Stukowski and Rajan, 2010- pending publication).Work by Kakuta et al. 14 have developed one of first published work in applying field ion microscopy (FIM) a precursor technique to the m

39、odern day atom probe, for high throughput screening applications in biological studies. Their approach was to use FIM to rapidly screen DNA sequencing. Their approach involved: fixation of one DNA molecule in stretched form under microscopic view, cutting and obtaining the specific part of DNA to be

40、 sequenced, before observing it in the Field Ion Microscope. In this method specific part of the DNA can be analyzed selectively. Because this method does not include electrophoresis, there is no limitation of the sample length. Another advantage of this method is that all the processes are done in

41、vacuum or gas phase suggesting that it could be faster than conventional liquid phase reactions. The overall speed of chemical reaction is generally regulated by the diffusion or mixture of the reagents as well as by the rate constants of the reaction themselves. In the former case, improving mixtur

42、e will accelerate the reaction speed. Therefore this method can be a high speed sequencer because diffusion coefficient in gas phase is significantly larger than that in liquid phase. A metal needle with finely pointed tip is used as a sample for FIM observation. Fig. (5) shows experimental setup fo

43、r electropolishing of metal wire. W and Au wires were used as substrate and DNA molecules were immobilized on the substrate.The work of Kakuta et al. with the FIM provides a good introduction to the challenges of identifying the chemistry of individual molecular fragments. This is where the value of

44、 the advances and transformation of the FIM into the modern day time-of-flight leap apt can be appreciated.Atom Probe TomographyCombinatorial Chemistry & High Throughput Screening, 2011, Vol. 14, No. 3 201The application of the atom probe to low conductivity systems necessitates the analysis of

45、well defined thin films and structures in order to quantify the complex emission profiles and informatics challenges. Organic self-assembled monolayers (SAMs) represent an ideal test bed for (1) elaboration of the experimental details associated with the application of atom probe imaging to organic

46、materials; and (2) elucidating fundamental questions in organic surface chemistry. Self-assembled monolayers containing thiol and silane anchoring groups have become a ubiquitous tool for surface functionalization and interfacial design. Construction with _-functionalized n-alkanethiols, HS-(CH2)n-X

47、, are particularly appealing in that these materials readily chemisorb on metals such as gold, silver and other materials to form densely packed and robust surface films with well-defined geometry and structure. As noted by Kelly et al. 15 there is only a limited amount of atom probe literature repo

48、rted on organic films and these have focused on conducting polymers and self assembled monolayers. For instance Kelly et al. 16 have studied conducting Poly(3-alkylthiophene)s (P3ATs) which contain a polythiophene backbone, by using a dip deposition approach on prefabricated atom probe tips.Nishikaw

49、a et al. 17 have conducted APT mass analysis of amino acids by depositing molecules among clumps of single-walled carbon nanotubes (SWCNTs), grown by the high-pressure carbon mono-oxide process (HiPCO). Densely packed SWCNT fibers enable production of specimens with ample quantities of biomolecules

50、drawn in via capillary action that avoid catalytic modification due to contact with a metal carrier. A ball of tangled long SWCNT fibers, several tens of micrometers in diameter, was silver pasted onto a tungsten tip and then dipped in a solution of sample molecules of glycine, cysteine, leucine, or

51、 methionine.QUANTIFICATION OF DATA: DEALING WITH THE DATA DELUGEWhile an APT is a powerful technique with the capacity to gather information containing hundreds of millions of atoms from a single specimen, the ability to effectively use this information creates significant challenges. The main techn

52、ological bottleneck lies in handling the extraordinarily large amounts of data in short periods of time (e.g. giga- and tera- bytes of data). The key to successful scientific applications of this technology in the future will require that handling, processing, and interpreting such data via informat

53、ics techniques be an integral part of the equipment and sample preparation aspects of APT. For example, 3-D spatial reconstruction of APT data will permit us to view approximately 10 to 50 million atoms. Data sets from single specimens will contain 200 to 300 plus million atoms; however, visualizati

54、on of 50 million atom positions are currently not possible. While the experimental data on organic materials is limited, the next step is to develop imaging data along with spectroscopy data. In this section we describe some of the approaches to address the issues in quantitative imaging and image r

55、econstruction that are critical to both organic and inorganic materials. While the following discussion will review the work in this area in metals (also a relatively new area of research) it represents a good guide to the problems in organic materials as well.There are numerous set of instrumental/

56、operational parameters that influence the spatial resolution/accuracy of each atom recorded in the final image. A good example of how instrumental parameters can affect the image is the case of the effect of laser energy in thermally pulsed evaporation. Cerezoa et al. 18 have shown for instance that

57、 at high laser powers, the spatial resolution between atoms appears to degrade (see for Fig. 8). They attributed this to surface diffusion effects that are reflected in loss of depth resolution.Images as shown in Fig. (7) are influenced by many factors and one of the aims of this project is to quant

58、itatively assess how ALL parameters ranging from operational issues of the instrument to attributes associated with specimen geometry and materials characteristics influence the image. Hence one may say that we will effectively extract the “digital signatures” of the influence of these attributes on

59、 the image; analogous to the contrast transfer function concept in high resolution transmission electron microscopy. Along with the materials characteristics outlined earlier, the instrumental attributes that we plan to study in a combinatorial fashion on each set of materials include: laser energy

60、(in laser mode), base temperature, pulse fraction (in voltage mode), pulse repetition rate, evaporation rate and specimen shank angle and tip radius. When combined with the different materials we then of course have extremely large and diverse, multidimensional data sets that form the experimental foundation for developing our