 Thank you for inviting me, Professor Tobias. Again, my name is Mitchell Brown. I'm the chemistry librarian. I also do earth system science, Russian studies. And for a certain amount of time, I'm the chemistry and the physics, astronomy, and mathematics librarian as well. So I do all the physical sciences. I do all the earth sciences. We had a colleague leave, so I'm taking over their position. What I want to do today is cover chemical information sources. What I want to do is talk a bit about what's available in the library. Not only individual databases, but the kinds of information you would find within them. Some of these are probably going to seem familiar. You may not have ever thought about using them in the ways that I'm going to demonstrate. And because our time is fairly compact, what I want to do is demonstrate how you might be able to answer a question. We'll look a little bit at the database, but we'll cover sort of a variety of different sources today. The interesting thing, hopefully, you'll take away is, these are tools. It's meant to be a toolkit of possibilities. A lot of the answers you can find might come from one or two sources, but there's always going to be a specific kind of question that you're going to need a different kind of tool. And it's good to know that those are available, even if you don't use them on a regular basis. So you don't need to use the same tool for every question, but there are some options for looking for other kinds of objects. So what we'll do is cover different kinds of material. Try to do some of the work, but we're not going to have a lot of time to spend the entire hour, hour and a half, on any one resource. Now I handed out some slips of paper with passwords on them. What we're going to try to do is spend a little bit of time with SyFinder Scholar, enough to get you started. These are teaching passwords, so they'll work. The accounts that you signed up for will also work. The reason I'm using teaching ones, it gives me another set of 40 or so available spots, so we can do it during class. Oops. So I'm going to cover sort of a mix of different databases, and what I want to do is break them down into different kinds of tools. So what we're going to be looking at are largely three different types of information sources. The first librarians like to call bibliographical. These are databases where you'd go to search for a topic. So I have a research idea in mind. I have a certain researcher or group that's been doing work. I want to find more papers like that. And this is probably the one that's most familiar to you. Academic search complete is an example of a database like this. SyFinder Scholar, we're going to be looking at that web of knowledge, web of science, is this kind of a database. I have an idea in mind. I want to find anybody who's written a paper on this topic. I want to find other papers like the one that I found. The reason we're interested in this is not only can we find information about an idea. We can look to see where that information was referenced. So what are the foundations for our understanding of knowledge? We can also look to see who's published subsequent to the article I have in hand. I want to look into the future of a research paper. Who has been working on this since it's been published? Which directions have they followed? Is there a divergence of opinion? Are other people doing similar things with the idea? Or have they taken it in a completely different direction? Second part of what we're going to be looking at is databases called citation. And that's the one I was talking about just now. What's happened since the time the paper has been published? Web of science has this particular characteristic. Increasingly journal publishers are including that in the work that they do. So American Chemical Society will refer readers to other papers of interest or other papers that refer to the one that you're looking at. Google scholars are also an example of a reference where you can find other papers that reference the one that you're looking at. And again, the reason you'd want to know that is you found an interesting paper and it's helpful in your work. You want to find more like that. One way of looking for it is to see if anybody else has used that as a reference and then look at their work. You can see a chain of development of an idea. The last source is probably one that's most familiar to you if you're doing lab work. It's a numerical databases. So if you're looking for chemical, physical property information, but also structural information, if you're looking for preparation or if you want to make an object, if you want to make a crystal, if you want to do synthetic chemistry and you want to find preparation strategies for particular compounds. Not only can you look for the availability of those kinds of strategies, but you can also place limits on what kind of synthetic pathway you want to find. Do you want to find it in one step? Do you want to find multiple step process? Are you looking for a preparation strategy that has a yield above 85% or you want to limit those to only the yields above 90% so it's possible to find those. You can look for exotic preparation strategies or something that will yield a product with limited number of steps, a certain number of reagents, and high probability of a single product outcome. One thing we're going to cover today, all of these databases will work on campus. If you're off campus, if you travel, if you're at home, you're going to want to use the VPN, Cisco VPN client to work with all of these products. So even though they're web based, particularly the SyFinder Scholar client really needs multiple windows to open. So you're going to want an environment where you have a database window, you have the journal article database open, a web VPN will only work to look at the journal part. It doesn't open multiple windows. So now we're just in a layer of technology where you're going to need multiple viewers. All of this will also work on your mobile devices. Increasingly, if you have iPads, if you have phones, smartphones, we're searching the databases on these things or beginning to be enhanced along these lines. SyFinder Scholar will allow you to search as long as you've registered for account anywhere, either on campus or off. The nice thing is you're guaranteed to connect off campus as long as you're registered through the accounts. It uses a completely different set of login procedures. The only thing you can't do is structure searching on your smartphone. And I hear a groan of that sometimes when I go to ACS meetings. Some people would like to do structure searching on their iPhone. It's just that big. So how do you get your, how do you do your work on that level? Having said that, people are still interested in doing work wherever they are. What we're going to do is start with the SyFinder Scholar demo. One thing I'd like you to do is from your computers. We switch over to one more. The library's website. So if you're starting from a main screen in the university's webpage, probably if you're landing here, you can click on a link for the libraries. What I want to do is click on the databases to get started. And under the science and technology databases, we have a link for chemistry guides. And we're going to click on the website, the SyFinder web login. Now just to show of hands, if anybody who's walked in late, if you don't have one of these pieces of paper with the passwords on it, just show of hands. You can pass some of those out. If you log in with these, go ahead and use the username and the password layer. I'm going to go ahead and log in under my account. So I'll cover some early developmental material, but we'll get back to the database. One thing I want to point out is that the chemistry library I'm responsible for explaining this. SyFinder for the use at the University of California is for educational use only. By accepting this boilerplate license, you acknowledge that you're a staff member, faculty, or a student. It's only meant for academic use. You can collect, you can store references, you can collect up to 5,000 records on your own personal accounts at any one time. If you reach above 5,000, you won't be able to store anything on the account. If you export those files and then store another 5,000, that's possible. You agree not to use any automated scripts against the database, and you agree not to share your username and password with anybody outside the university. Having said that, I encourage you to use the account rigorously and fully. There's a lot of really interesting work to be done within the database, and it's got some features that are both easy to work with, but also it's a tremendous tool. It's a kind of database that you wish you could have access after you've graduated. Unfortunately, your access will drop once you've graduated from the university. And you might see this script the first time you ever open up SciFinder on your browser, whatever browser you're using. There's a Java app that installs, and that's for the structure drawing elements. We're not going to do much of that today, just because it's a slightly more intensive draw on the servers, but I'm going to do a brief demonstration of that. So I'm going to get back to this in just a moment. I want to do a little bit of explanation of SciFinder as the database. It is the chemical information system for the Americas. It also includes global and international information. So the coverage goes back to the early 1900s, and it's based on a couple of different systems. One of them is the bibliographic part. So we're looking for papers. We're looking for research. The other part, the part that makes it really functional as a chemistry system is a registry number-based identification. And this is a big deal. You can find not only a compound that's specifically for what you're looking for, but you can also find variations on that compound. So for instance, let's say I want to find both an element, a particular substance. I want to find a chiral component of that. I want to find one that substitutes a deuterium for a hydrogen. Each of those will have a separate number. So it allows you to separate not only the identification of one object, but anything that's related to it and anything that falls into a class of compounds that are similar. The reason we do this is there's roughly 27 to 54 million, depending on which numbers we look at, but 27 plus million organic inorganic substances. They've also started adding sequences for enzymes and other kinds of biological sequencing into a database. So in order to distinguish the differences between a substance, but maybe one that has an ionic form or a substance and one that's a salt form of that compound, there's a different number for it. Within the database, in addition to chemistry, there's also engineering materials information. So you can find coordinate systems, combined coordination compounds. So add mixtures of material that's together, alloys, polymers, salts, gene, and enzyme sequencing. And the part that we're going to be looking at in addition to just the bibliographic information, physical properties, experimental properties, and also calculated properties. So in addition to any experimental information that might be derived from the published research, increasingly chemistry databases used to calculate properties of compounds. So if you're doing lots of research, you're not necessarily doing lab work on every object, but you can calculate sort of estimations what might be certain chemical physical properties. And the database allows you to break down into certain structure and information debate related searching, depending on the kind of work that you're looking for. So we can find information about chemical structures. We can look at individual substructures, so we can look at anything that has a certain skeleton form. If we're looking for similarities, if we're looking for compounds that may have some of the same structural elements, but could be part of a polymer, or something that has maybe an inverted structure, so it has some of the same elemental empirical formulation, but has a different stereosculture. We can also look at reaction information. Again, we want to find preparation strategies. We want to find what techniques, or what kind of multi-step reactions there might be. We can look to see which part of the reaction steps one of the compounds falls into. Most often we're looking for research topic, or we're looking for a particular author's name. The substance identifier information, this is the registry number. It's used for identifying not only individual compounds, but also preparations related to that compound. And then if we're looking for chemical or physical property information, it makes your search much easier if you have the registry number, then you can find that compound specifically. If you do a, if you try to do a commercial name source, or if you try to do an empirical search, if you're looking just for physical properties, you might have to feed through maybe several different variations of names or compounds. So one thing that can help, if you ever do identification of compounds in your own research, try to see if you can find a registry number and add that to your notes. It can end up saving you a lot of time when you need to go back and either look for more information about that compound, or if you're looking for things related to it. So if we switch back to SyFinder, the first thing I want to point out is the interface can seem deceptively simple. It's built for practicing chemists who don't do a lot of their own researching themselves. So it's meant to be easy to work with, fairly button-free, fairly straightforward, but behind the scenes is a really robust, very well-thought-out research strategy. So instead of doing searching like you might do with Google where you simply throw words on the screen, SyFinder's got a different strategy. They call it a natural language approach to asking questions. So what they would like you to do is phrase your question in terms of a real sentence. So I might be looking for a particular example, and what I'll use is the details off of the main screen. So instead of just saying noun, noun, noun, you know, sort of breaking things together into parentheses, what they would really like you to do is simply put in a phrase. So I want to look for the photocyanation of aromatic compounds. And if you're looking for a particular kind of aromatic compounds, you could actually just include the adjective for that as well. Or the first one, the effect of antibiotic residues on dairy products. And the reason the database does that is it's meant to help researchers who don't necessarily know not only how their compound could be described in both simple and more complicated ways, but it's also open to the possibility that not only do I want to find exactly what I looked for, but I also want the database to help me find other things that might be interesting to read as well. So let me show an example of what this kind of search might look like. Now I'm going to run a search based on this. I want to point out that like a lot of other databases, we've got some features that can allow us to narrow to the kinds of material that we're going to find. So in addition to finding things like journal articles, patent information, there's clinical trial information, there's conference papers, there's letters to the editor, dissertation information, technical reports, written in English primarily, but it's an international database. So what we'll find are abstracts are written in English. The original papers will be in whatever language that we're published. We'll actually see that when we come through the sets, but if you ever find you're searching through results but come back with language as maybe a challenge, you can always filter at a certain set. And unlike some databases where you only choose one and then none of them are available, in this case I could say English is acceptable, I can read French and German so I'll take those two and I'll also include Russian, but I'd rather not get Chinese and Japanese. So I can choose maybe a set of the kinds of documents I want. We can also narrow down to certain time frames and occasionally you can also narrow to a corporate entity. So not only can you find a person, an author, but if you knew they were working for a particular research group or a particular company, you can include that as a limiter. And one of the reasons for doing that is let's say you're working later on in your career for an industrial concern, so you have a commercial interest. You want to find what your competitors are doing. So you want to see if the ideas that you have, if anyone's published from 3M or if anybody who's working with Amgen has done work in this field. And it could be we're not necessarily spying on them, we're just interested to see what kind of research they've published. So I want to know what my group is doing. I also want to know what another group is doing. And this is another way of narrowing down to some of these fields. So let me go back to the search again. What I'm going to do is enter the term as part of a sentence. It doesn't have to be a fully defined sentence with capitalization and periods, but a phrase helps. And what the database will do is break up the question and group it in terms of structures within the sentence group. So in this case it found two references that contained everything that I was looking for. And then it breaks down individual words or groupings of words. So originally the question was aromatic compounds. It'll treat that as a phrase. And any time either the concept of what I was looking for or maybe variations in that term are used. And an interesting thing that happens is in addition to the word that I used, there's a thesaurus that might go through the back and look at aromatic as a term and come back with things like volatile organics or different variations of things that are organic compounds or aromatic compounds. And that's how this number can be quite large towards the end. Some of these are based on a relevance ranking. So the further down the list you go, the larger the sets will become. They include anything that's above that. So if I choose 26 references here for these two concepts, it includes the 15 and the two above. So in this way I could look for something that's very specific to my question or something where the neighborhood's a little bit larger. And the interesting thing is sometimes it'll come up with alternate words for a concept that I didn't necessarily think of. An example is I was working with a person who had a question about doing analysis on seawater. And one of the terms she was looking for was either spectroscopy or some way of evaluating the turbidity of a sample. So turbidity is the definition of how clouded or, yeah, cloudy, that's a good sort of broad term. A seawater sample might be. And so the chemist in me is thinking what kind of things could be part of that sample that make it whatever, cloudy. When we did the search, in addition to turbidity and turbulence and other related forms, the word brackish was appearing. So a brackish sample is a cloudy sample. It's either a lot of suspended solids, maybe suspended organics, different kinds of chemicals within the water. Certainly brackish water is cloudy water, which is also turbid water. So I didn't necessarily look for the term brackish. The database interpreted that and helped me find other papers like the ones we were looking for. So the database will highlight any term that pops up in our search. It will also tell us the source for the material, so the actual reference to the papers will appear. Again, we're interested in English language documents. It'll tell us if it's English from this level. We can do some analysis and refining of work. Largely, we're interested in the full text. When we look at the full record, not only can we find, you know, the paper itself, what chem abstracts will do will actually break down within the body of the paper. Any compounds that are identified as part of the research. So if I'm looking for concepts, that's one way of looking for similar related topics, we can also see what chemicals, what physical structures are used within the research. Any of these are searchable. So if I was looking specifically for maybe the properties or the calculations of, in this case, nitronathaline or different carbonyl forms that can click on these links, you notice a little bit further down, there's a set of registry numbers with a P that follows them. Now within the database, it'll identify the structure of a particular compound. Any time the record indicates a preparation was used, the P will follow. So if you were looking for preparations for a particular kind of compound, this is one way of identifying them within the search. There's also another term that's used, a DP, so derivative preparation or derivative compound. Those are identified. That's sort of a carryover from the print guides, but it's a way of identifying within the paper, did they use these as part of a reagent or as part of the process or were they really making something? And you can actually click on these and find other papers that share some of those characteristics. So how do I make this compound? The paper talked about that, what did they use? Since it is a research database, what we're looking for is the full text and across the top there's a series of buttons. We can look to see who the, we can look at the references that were included in this paper. We can also look forward again to see who else might have written a paper that referred to this. And then the full text button, if we click on that, it'll link to the full text if it's available. So usually the button is an orange, you see E-links button. In this case, you know, it's a different color, it does the same thing. So it'll link out to full text where it's available. In this case, this publication is a monograph, so it actually is in AMP-PAC. It's something where we have to physically get the hard copy. If we were looking at the brief record, we can also look for the full text dramatically right off of the short index field. Now, we only have two references here. If I go back and I look at one of the slightly larger sets, the database allows us to filter and analyze the results off to the right side. So if we were looking for a topic, and the paper seemed interesting, but we're also wanting to know who's publishing and from where, we can look to see which authors might be the most active in the field. If we wanted to limit to other kinds of features, you know, what particular topic is being discussed? Is there a journal where these ideas get published more often? Potentially, I have a paper I want to publish similar to these ideas. What's the best journal for that? If I'm looking for publication year, when were these published? Has it been something recent? Is it something that might be older? Is there a company that might be involved in this kind of work? So what's the sector am I involved with? Now, 71 is a fairly moderately sized set. Maybe I want to narrow that down a bit. So I can narrow from this 71 down to smaller sets. And again, looking either at specific ideas or just adding another term, another research topic term. So as a research database, SciFinder has not only the common literature, the published literature, it looks for any time a compound is ever identified in patents and commercial publications and theses. So if you're looking to do the most thorough, most involved research, this is definitely the database for that. But it might be a bit more complicated than simply doing a weekly update. If I just wanted to find out what's published, it can be a little harder to get into sometimes. And maybe it's got too many tools for it. So it's helpful and it's useful. It's not necessarily something you have to use as the default search for everything that you're doing. But it does have some characteristics that make it interesting to work with. And one of these is the structure searching. So let me ask you not to launch the structure search. You can look at the tool. Please don't run the search. I just want to run one brief search just to demonstrate how it works. Otherwise, it's going to actually pull down the reaction from Ohio. So the web-based drawing software is similar to ChemDraw, similar to Marvin Sketch. It's got some of the same drawing characteristics. It allows you to go in and draw the structure if you like. Another nice thing is available if you've already pre-drawn a compound. Like say you've taken the time, you've done some work ahead of time as I have. I wanted to pull in a structure because I didn't necessarily want to sit down and redraw something. I just want to be able to pull up an object that I've already drawn and do that as a search. So I did this in ChemDraw. What I did is imported it directly into the database so I could do the search. Now, this is not a fairly, this is not a really complicated structure. I probably could have drawn it out. But if you've got something that's maybe a bit more involved, you can take some time to draw that offline. So I imported the object. I can add some identification if I like. But then I'm going to do a structure search. And what I've got here are some options. I can look for an exact search. So just what I've drawn here, sometimes you have sort of substitutions if I'm looking for a connection site. I can include that. A substructure search will look at an element of this, usually the ring structure or any of the connection points. And then look for any compounds that share some of that and more. And then the last one's termed a similarity search. So it may not actually have the same connection elements. Maybe they'll be in an inverted form. So the methyl and the hydroxide will be switched. With a more complicated structure, chiral variations and anantomers will also be included. You could also have certain compounds where a lot of it's here, maybe 80% of it's similar, 90% of it's similar, but there's a different substitution somewhere. And for people doing things like drug discovery or synthetic chemistry, looking for close cousins or things that are a little bit different, is an interesting research strategy. In this case, I just want to do an exact search because it's fairly quick. So I'll pull this up and then allow me to look for clarifications of some of the characteristics. In this case, I'm going to narrow to only single components. I'm not looking for anything that's multi-component. I'd like to find a preparation strategy and any time this might be used as a reagent or a reactant. It came back with a fairly quick result. It's got three different versions. It tells me how many papers are related to each of these individual records. You'll notice there's a different registry number for each one, and now comes a question. If I did an exact search, why isn't there just one result on the screen? If you look at these three, is there something about them that's similar, but is there something that's different? So why do I have three instead of just one? Any thoughts, anything you see on the screen that might be an indication? So in this case, this one is an ion form. It's structurally similar, but it's got a different electric charge to it. And the last one, certainly this element is correct. It's actually part of a polymer. So similar, yes. Structurally different, definitely. So even though you're doing a database search and you think you've asked a very specific question, you'll still be asked to make some judgments based on what you're looking for. And there's a time and a place where you may want these different variations. In this case, we're just looking for this compound. And that one specific registry number will help us just find this compound whenever we want to go back to it. So let me show you some of the other features that are here. We can find more about this compound. We can look at the physical properties. There's also spectra information that's available. Depending on the compound we're looking for and its availability. So again, we were hoping to find preparation information. Clearly that's involved. Some other kinds of information are available. Physical properties, uses. I'm looking for biological involvements. This is roughly a salicylate. So it's used as an analgesic. It's a family of painkillers. It's part of that profan family. So a variation on different kinds of compounds like that. The physical property information will be included. The citation for the source will be listed off to the right. Some of these are identified as physical properties that might have been measured. Some of them are calculated. If we go down to the very bottom of the screen where the references are, 22. We'll see here where the references might be included as individual papers where they were looking at a compound. One here, FISPROP data was identified from a particular research database. Here we go. AIST integrated spectral compounds. So some of the spectral data was coming from research groups. ACD, sometimes will be appearing. It's a chemistry software that's used to calculate certain physical chemical structures. Let me just point out one of the spectral references. So for this case, this is an infrared spectra. If you're looking for information on this, it's something that would show right up in the database. Not every compound has the same amount of coverage for physical chemical properties. So you may find that some of this is, the coverage is not necessarily the same for every compound. On the other hand, it tends to be material that's either used largely in other processes or has a very specific biological use that would suggest more closely discovered information about the physical properties. Normally, I don't come to SciFinder if I'm looking for melting point, boiling point. I do come to it when I'm looking for something that might be a little harder to find. So electrical properties, if I'm looking for spectra information, if I'm looking for maybe mass solubility, some other kinds of physical values that aren't necessarily available in easy-to-find handbooks. But again, a resource that's helpful depending on the level and rigor of your question and depending on the ability of you to identify a compound specifically, it can be a lot of help. It's very useful, very quick, really straightforward to use. One reason we ask you to sign up for individual accounts is you have the option of saving these results and maybe coming back and looking at them later. So up on the far right-hand side, you've got the option to be able to save these results. So you don't have to type this in again. So you can save both your search strategy but also the results that you found in those papers. And let's say you're doing some work within a small group. Maybe it's a lab group, maybe you're doing some research with some colleagues. You can save that search set. You can also share that with other people within the UC system. So if you have other colleagues at Irvine, what you do is send an email to their email address, they'll be able to log in, you can share that account from your account to theirs. They would actually get the live documents. So let's say you found a structure, physical property information, maybe preparation strategy. You can share that with another colleague. They pull it up in their account and they can work with it live. So you don't have to re-key the search. You don't have to translate it from paper back into electronic again. You can actually keep it live and work with it on an ongoing basis. Unfortunately, the one thing it doesn't do, you can import active structures in a ChemDraw format. If you export, it tends to reduce it back to a flat file like a JPEG or a GIF. So unfortunately, the structural parts aren't included. It's sort of half and half. It's a way of, site finder's way of controlling that their database doesn't necessarily become something that people just download to get the structural active parts of the chemistry. Having said that, it's a really functional, very active database and I recommend people use it as much as possible as students to get aware of the details that it can do. What I'm going to do now is log off because I want to move on to some of the other information that we've got. So now we're going to move on to a different kind of a database. A site finder has some of these characteristics. What I want to do is look at Web of Science and I want to compare that a little bit to Google Scholar. When we're looking not only for bibliographic information, I want to find a topic. I want to find a person. I want to maybe even find a physical value. But I also want to see if anything's happened since it's been published or I want to learn a little bit more about that paper based on the references that it used. It's also a really interesting way of coming to a topic as somebody who's not a specialist. So say you have an opportunity to work in a research group. You want to find out what's been done by that group but you also want to learn a bit more about the topic that they're working on. And the hardest thing is usually where do I get started and how do I understand a topic where I understand parts of it but I don't understand where it came from and I don't necessarily understand how the work that we're doing here relates to the work done in other places. So I want to learn a bit more about how we came to understand the material that we're working on. So textbooks are an interesting way of thinking about that but I already know the fundamentals. I could have a talk with people in the lab and that could be really helpful but sometimes people are busy and they're not really available to sit down and sort of tell you this is what we're doing. This is the shape of the research that I'm working on. So instead database like Web of Science can help and let me point out some ways of doing that. Web of Science includes not only the science literature which we're most interested in. It's got arts and humanities. It's got some social science information. We may not be interested as chemists in doing that kind of work. Occasionally you know the social implications of science will pop up so when we're doing climate research arts and humanities social science occasionally comes into it but largely what we're looking at is research literature across the world. It's not as deep as SciFinder. So if we wanted to find everything published SciFinder is the place to go. Web of Science looks at the top part of a literature structure. So if we think of it as literature is a mountain. SciFinder is going to be maybe the top third. I'm sorry SciFinder is the whole mountain. Web of Science is maybe that top third. So 8,700 papers maybe 20% of the world's publication but out of that 20% about 80% of what's ever published and referenced comes out of that top part. So they're commercial publishers. There's the society publishers. The material that's most often distributed and most widely read. That bottom 80% will be individual journals that might focus on a very, very small very specific set of the research strategy, you know. Like a journal that only publishes on orchids as opposed to maybe the Journal of the American Botanical Society or Proceedings of National Academy of Science that also includes botany but includes other topics as well. So Web of Science covers the largest part of the literature but the most active part. So like a lot of databases we're looking for a paper but within that paper there are things we can do to narrow that. We can look at the cited reference searching. So I write a paper. I refer to 14 other papers as source material. I want to fully attribute any idea that's not my own to the original source. I can show people this is where the previous idea, the foundations from my work comes from. That's helpful because I might refer to say a handbook for physical properties instead of reproducing all of the data I just refer to an element from that. It's an opportunity for other people to go back and look at that source material potentially for other kinds of work that they're looking for. But then I'm also looking at forward citations. The work is published in 1990. Has anybody worked on it since then? And if so, maybe they refer to my work. Let's see what they've done with it since then. So you're looking to see anything that's gone into the future of your document. So some of the things we're going to be looking at, we can look at references. We can look to see who cited the work. And then we can look to see potentially what a particular author has worked on. Within the database we can look to see not only how many times a particular paper was cited. And I just chose this one because the record was fairly compact. So since it's been published how many times has somebody else subsequently referred to it? They cited 88 references. I can look at each of those individually. And because the database may actually link to them, I might be able to follow the link from within the database. So I found a paper. I can read its reference. At the same time I'm reading that reference, I can also find what it references. So instead of what we used to do, find a paper, open up the journal, read it, look for a reference, pull the other journal down, read that. Now we can do this electronically just by following this thread of connectivity. An interesting thing that's happening is this concept of related record searching. And this is where cited works become really interesting. So I find a paper that's of value. The authorship is strong. The subject material is coherent. The publishing, the publications credible. All the things that I like about a paper. I did a keyword search and I found it. That was good. But I want to find other papers that are related to it. This is where the database searching and librarians come in. So what we've done is connected to other papers that are related to the one that you're looking at, but based either on conceptual ideas. So what area was the paper focusing on? What are some of the broad themes within the paper? Or if you cited a certain number of papers, we look at other papers that share some of those cited works. And the argument is if we both use foundations that are similar to each other, maybe there's a relationship between those papers that might not be reflected in words that appear in the abstract. So if I'm doing work, maybe spectral studies on a particular compound, Professor DeBias is doing work but in atmospheric chemistry where he's looking at particular class of compounds. He wants to use a technique. He adopts my technique in his work. They're not necessarily connected except by that technique. I might find other papers that share that similar technique but it could be used in a completely different area that I never thought of. And the reason this is interesting is I might write a paper where I study just the ability of my particular analysis to work well at identifying certain kinds of compounds. But what people are really interested in is can I detect this compound if I use a strategy that's unknown? And my paper could give them that strategy. And this is how we find things that might be helpful but we wouldn't necessarily have found them if we did just a keyword search. So what I'm going to do is pop back to the chemistry guide if we hadn't closed that. What I have is a list of important sort of commonly used databases including Web of Science. And just a quick show of hands. How many people have used Web of Science maybe in writing 39C or other classes? Just a quick show of hands. Somebody? This is probably going to be your favorite database because it works really well for most of the questions you're ever going to have. And the reason it does that is it covers not only chemistry but also related physical science topics. Interface is really easy and it's the most common popular publications. So we're looking at the top of the literature. It's also easy to get into and you're more likely to use it first before you switch into SciFinder. And just because I want to do an interesting demonstration, I'm going to pull up a very specific kind of search. This is something I want to point out because it will work for other kinds of databases. Most word, most searches use an implied and. So if you just put in a string of words, the implication is always the word and appears between each of them. If you ever want words to appear next to each other's of phrase, use either parentheses or quotation marks. You can use this for Google Scholar as well. But a way of encouraging proximity. There's also a truncation feature. So if you want to find simple plurals or if you're looking for maybe a compound word, use an asterisk key. This will give you the opportunity to search not only the word that you're looking for, but maybe related terms that are like that. If you do this, sort of mix these together, you can get a search that maybe pulls in more records that then you can narrow. This idea of doing research is about an inverted triangle. So if we talked about literature as being like a mountain, your search results should be more like an inverted mountain. I prefer to start with something more and narrow down to the better. So we're looking to remove the dross and just find the gold. And the way of doing that is give yourself enough to start with so then you can narrow to something more specific. Very rarely do I ever hit it right on the button with the very first search unless we put in something that's really, really, really specific. So I wanted to pull up one record and use that as a specific example. So here's an example if I threw a lot of keywords at a topic, not only did I find it in the title where I expected it, it will show up anywhere within the body of the abstract. So the more opportunities for words to be found, the more likely something will come up. Increasingly, what we'll find is the author information will either be sort of highlighted or it'll be fully expressed. We can look to see what kind of subjects might have been covered. In this case, we can look at the keywords or the large groupings. We can narrow to certain locations. Of course, I was doing this specifically. So you knew it was coming from a university site. We can limit to a particular organization. This is an example of where we can find the full text by using UCE links. What I wanted to do is click on the find related record search. And what I'm looking for are papers that share some of these 88-sided references. So we're starting with the first paper. That's our base. We found other papers that the second one had 74 references. The first one had 88. But out of the shared combination, there's 23 of them that are similar. You could ask yourself, well, why is this interesting? A lot of times you're reading individual research papers and it's an idea. And it's maybe very, very specific, but it's a good idea. You're not necessarily sure if I read another research paper or how they connect to each other. Largely, there's some broad idea usually. There's some central idea that's important and most relevant. So the central core, that most significant idea. And then other papers might highlight that or give an example of how it connects with another idea. Not every paper is unique and special on its own. It's not their fault. It's not that everybody comes up with the best brand new idea every time they publish a paper. So you'll find this interesting paper. Other people connect to it and either describe it more fully or give an example of their own. When you read some of these shared references, you'll start seeing, oh, such and such referenced that paper as well because it had this sort of big connecting idea. If I read that paper, the next time I see another work that refers to it, I'll think, I understand where that came from so I don't necessarily have to reread it again. What you're doing is creating a knowledge map for yourself, understanding not only how these papers relate to each other, but what's the central theme underlying them. So if we just click on those 23, we can actually see what they are and a little bit further down. Increasingly, with the databases being integrated, potentially we could actually follow these through and we could look at them individually. Occasionally, you'll run across topics where the references might not be linked. So this is an evolving technology. We started out with journal literature, trying to find a way to make the journals available electronically. Books weren't available in electronic forms when we started this back in 92, at least not a lot of them. Conference proceedings were sort of half and half. Now we're 15, 20 years in, we're starting to work with linking to books as well. So over the next five to 10 years, we're going to see more of this kind of linking gel. So the more you use it, the more likely it is to connect. Right now, we sort of come up pretty close and then you have to use the print guides for a lot of the other literature. It's not a bad way of thinking about how the world has changed. It just means we haven't got there all in the same jump. What I did is went back to the related record search because I wanted to show you out of a fairly large search strategy of almost 45,000 records, some ways of narrowing these down, maybe a little bit more compact. So if you remember, I mentioned if you were in a graduate student situation or brand new to a topic, you might want some way of understanding the idea, you know, give me a way of sort of gelling this a little faster. So one thing I recommend grad students do is let's say you run across a topic or this really big research idea, try to narrow it to review articles. So this is not a book review, it's not a mover review, it's an academic review. And what these will do is give you a couple of things. One, it's going to cover a certain area of knowledge, so a topic of the paper will be maybe a 5, 10, 15 year review of that art, you know, what is the state of the art of this science? It's going to usually have a whole bunch of references, you know, several hundred references sometimes. Yeah, here we go. So a paper that covers the process of looking at spectroscopy, 111 papers. So in a sense it's done a research for you, came up with a list of 100 papers based on spectroscopy across a period of time. They're also going to talk about how they relate to each other. Now the authors might have an internal bias, which is not a bad thing, but if you're doing, you know, infrared raymonics spectroscopy, you're going to want to focus, you're probably going to focus on that, so you're not going to cover everything, other kinds of spectra techniques. But that's not a bad thing. It's going to focus on the areas that you might know. Ideally, if it's well written, and most of the time they sort of fall into this category, they'll describe how the ideas are similar or different, and they'll start grouping together people and research and techniques and explain how it has come to this point that this is our understanding of this technique. There are other areas that might move in different directions, and this is a change that was made. And largely there are decision trees that come up when you're doing research. It's a good idea, but which one worked best? You know, this one worked, but it requires the technology. It was a little complicated to use. So somebody did it this way, and it's more popular because it's easier to work with. We didn't forget the first side. Just a lot of people are doing this. It could be we fix the technical issues on the left side and we keep doing that again. It doesn't necessarily mean they're better or worse. They worked or they didn't work. Sometimes they'll talk about things that were just dead ends. You know, we did work, found some interesting things. It just didn't go very far, or we answered what we were looking for, and we stopped, and then we went in a different direction. It's a way of learning a bit more about a topic without having to eventually read each of these individually to understand that. Now, if you're doing the research in the area, you may end up reading all of those at some point. But it gives you an idea of where to start first. The other idea is about cited reference searching is, at first we were looking for a certain kind of paper with certain characteristics, the most recent materials published. So, if I'm looking for something that's published lately. The interesting thing about cited works is we can also look to see which ones might have more impact, or may have been used more extensively, which suggests potentially there's a greater value to them. All of these are sort of broadly described because just because it has a large citation count, it doesn't mean it's the best paper in the field. It just means a lot of people have referred to it. So, maybe there's something that it's worth looking at. Now, in general, the average citation count for papers within the database is about 1, 1.1. So, paper gets referenced maybe one time. Anything above 10 starts to be significant. Anything above 100 is identified as a different kind of level, more significant. So, several hundred, several thousand is clearly higher than the average. Clearly, these might be interesting because they cover really big topic areas. Maybe individually, either original techniques or the first time an idea was presented, or a strong reason or a strong indicator that there's something worth looking at. Looking at these individually, it's hard to know. If you can take these papers and then ask somebody who's in the field, is this a good example of a paper? And they'll say, oh, sure, then this is why it's important to read, and if you like this paper, you can find more like it. And the question is, how do I find more like it? Again, we do a keyword search, or we look at the papers that cited it, so we look to see anybody else's writing in that area, or we do the related record searching, which is a lot like doing keyword searching, but it's also conceptual. So, I start with a paper, I find other papers like it, find another interesting paper, I do the search again. After a while, you'll start seeing the same papers repeated, which is good, your neighborhood will become known to you. The neighborhood becomes familiar. Another reason I like Web of Science is it's easy to both look at subjects, but you can also narrow it down to individual authors, or if you're looking maybe at the history of a publication, history of people publishing on a topic. So, I want to pull up an example since we're here. If you're working with a research group, or if you're working with a faculty member, and you're interested in knowing what the group has done, what we can do is search for individual people, and then we can narrow it to a location. The database is a European base, it doesn't fully express the first name, the middle name of people, so it uses first and middle initials, that's why I'm using the asterisk. What I can do, though, is narrow it to an address. I would use University of California, but that can be abbreviated several different ways. I think the total within the database is about 14 different variations just on University of California. Ravine's pretty specific, so Berkeley's pretty specific, San Diego specific. You can also use zip code. So, let's say I'm looking for somebody, maybe the research triangle in North Carolina. I don't know where specifically in that area, but maybe somewhere close. You can use part of the zip code, so here I might use 926 and an asterisk. So, it could be somewhere around the campus, Orange County in that area. This is usually if you're doing business-related work and you want to find people who might be related, or to the university, but in their own research center, nearby, working on a topic. I do that a bit more for the folks who are doing engineering kind of work. Again, we can look to see what's published, how often, what kind of topics. It's very nicely cited work. A recent material. Again, the first one is the original example I used for the search strategy. So, the database is updated weekly, so we can see things that have happened currently. They go back further as well. Of course, if you've worked at several institutions or at subsequent work, we would have to change the address to try to follow that. So, one thing we can try is just doing a search based on the author, and the original 187 now comes 243. We just have to make sure it's the same person throughout. If you do a search by my name, I don't show up because my work isn't included in the database, but there are MC Browns that are not me that show up in the database as well. We just have to filter that to make sure we're talking about the right people. But, it's a very helpful, very responsive database. UCE links, targets, University of California holdings, which tend to be very strong, very robust. We just don't have everything that's ever published, so we don't have everything. Our new library alone is able to get things that we don't actually hold. Now, this comes up because it's worth comparing what Google Scholar is doing. You can search Google Scholar and you get results, so it definitely works well. The real question is, why would you do this instead of other databases? So, from a librarian's perspective, as long as you're on campus, anything we currently own if it authenticates properly and it works, great. It lacks some of the specificity of the other databases. So, if I want more control, if I just want to narrow it to certain fields, if I want to make sure I can identify the relationship between an author and location, maybe that works. On the other hand, if I don't know exactly what I'm looking for and I'm willing to throw a big net and just look at whatever comes back, you know, it does what it does. And the reason it works well is not only does it collect a lot of material from other sources, the journal publishers that we have subscriptions to open up their holdings to Google because they find anybody based on search and anybody looking for search will end up finding their material and just drive business back to it again. And that's why it works better than, you know, it wasn't built to do this, it just does it really well because the targets are available. So, here's something interesting that shows up. We find the original paper we were hoping to find, definitely. We can find related articles, four different versions. So, the question is, which of these is the original article? When we're looking at SciFinder, if we're looking at Web of Science, they'll target the publication source. So, the publisher itself. Increasingly, what's available for the authors to do, including faculty at University of California, is to either post-preprints or post-prints of their work on their personal websites or institutional repositories, which has been done here and here, also in this place too. It could also be not only the PDF, but maybe just the abstract of the paper as well. So, it could be it's available from a publisher's site. So, that requires a subscription. The full paper is available from the author's sites as well. That may not have a subscription. And depending on the paper that you're looking for, maybe you're just looking for a data element. Or maybe you want to look at it. It doesn't really matter if it's the fully formatted final version, you're just looking for a paper to see if it connects with what you're looking for. So, you may find it more than once, or more than one site. Another thing that Google scholars started to do is, this was a site click. So, you found the paper, you want to be able to use it and refer to it again. They've started doing this that other databases do as well. Come up with a representation that covers most of the details that you'll need for the paper. So, you can refer to it in your own work. Here's a caveat. It's a machine-driven extraction. It has a lot of what you need. It's not necessarily complete. So, what's missing here is the data publication, so the year, also any kind of volume or paging information. So, you need to know what the pages might be, or DOI, or some other way of identifying it. So, it gets you close, if not everything. Now, you notice off to the left-hand side, if you're off campus and you're not using VPN, this part will go away. You won't see that. We have our UCE links activated on Google Scholar as well. So, you can click on those and ideally click and link to the full text for whatever we actually own access to. It's a short answer. Google Scholar is an amazing thing. Sometimes it comes back with exactly what you're looking for. It also finds whether or not other people ever refer to the work, so they do a citation count as well. It's just not as easy to control. And sometimes you get stuff back that's exactly what you're hoping for, and other times it just seems to be stuff. I think search and discovery is something they'll be able to fix, and certainly they're working on it. And as a librarian, I really hope they keep working on it. So, I use it all the time too. I use it when I'm trying to check for a reference. If I'm looking for something, I don't necessarily know where to look, and I want to get lucky. I just don't use it as my only tool. So, I pull it out when I need it. It's just not the only tool in my toolbox. Okay, I've got about 10 minutes left, but I want to maybe reduce this a little bit smaller. The last is a bit more straightforward in terms of a database. It's a numerical information. I'm looking for a particular property. I want to find a spectra. I want to find melting point, boiling point. We've got a compound, and I need enthalpy to vaporization. Where do I look? Traditional sense was books on a shelf. We've done a lot of this electronically now. We have databases that link together handbooks. One of them is called Novel. Weird name for an interesting product. They search the details within the handbooks themselves. So, I can literally type in, I want heat of vaporization for Nathaline, or yeah, Nathaline. It'll look through handbooks, try to identify those targets. NIST, chemistry web book, does a lot of that too. Chemical physical, ionic properties, free information, some of it's subscription based. CRC handbook for chemistry and physics. Traditional print handbook, but also available electronically. Chemnet base, collection of other kinds of handbooks. Again, the idea that you could search just for a particular kind of compound. So, from the chemistry page, I've got those linked here. Let me point you to Novel. And again, what looks like a fairly simple interface, covers up a more complicated background area. You can also search using registry numbers, if you're using commercial names for a product, if you're using generic names. What it'll try to do is look for the occurrence within publications. It'll rank these based on the likelihood that it comes close to what you're asking for. These are not necessarily just PDF documents, but they're actually interactive. So, this one has both an interactive formula table, but it also has a way of pulling up the details specifically and looking at the calculations. So, we used to look at the handbooks, look at a guide, look at the table, come up with a value. What it is, it clicked on an equation editor. So, let's say I've got experimental data from my laboratory. I can actually put in the X, Y values, the X values here. It'll calculate how Y would follow. So, let's say I did, so an experimental value in lab. You've done the lab, you've done your work. I would type in the value or I did a measurement and it would drop it across the graph where those points are calculated and give the X, Y value off the graph. But I've got it in units. So, this one's in a temperature unit for Kelvin. If I switch this over to Celsius, this may have gone kind of quickly. The graph actually changed. The X, Y values, when I switch the temperature values here, maybe I want to change the function value from kilojoules per mole to calories per mole. And again, it does the readjustment on the fly for you. It allows you to export this if you want to decide it in your laboratory work. It allows you to export it. So, we've gone to a point where we physically used to look at handbooks to get values, to now being able to use them like interactive logging tools. Interesting thing about the values as well. I put this in, I was looking for a particular value. I can also narrow. So, let's say I had a boiling point. And out of all of these nathaline products, I wanted a boiling point that was either above or below a certain value. So, let's say 150, that's a molecular weight. So, let's do that. What I was trying to do is put in, let me put in this. Sorry, doing this on the fly is sometimes a bit much. You can filter within a certain range from molecular weight, maybe other physical value information to narrow to certain candidates that might fall. So, a critical temperature within a range. Actually, it also allows us to select the rows but also change the orientation of the table. Like I say, maybe all you wanted was molecular weight, critical temperature, and then the minimum max temperature values, compress the table just to the data elements that you're interested in and then just export those. Since originally we were asking for heat of vaporization, that column is highlighted, actually pulls up those values and allows them to look at it. If we look at the individual elements, it's an element on the original print guide. So, it's a combination of the full text for some books where you may actually want to read an entire chapter on a topic, but also tables, graphs, interactive charts where you're just looking for an individual numerical value and that allows you not only to search for that value but maybe an entire range. So, a compound that exhibits certain characteristics, give me back those possibilities and let me narrow that down to other sets of characteristics. Other kinds of handbooks we have, sort of sorry for the list of other ones, Merck Index, CRC Handbook, Physical Property, Chemical Information, Dictionary Sets, a lot of these will sort of be grouped together on that list of compounds we have off of the chemistry database. And the reason why it seems to be sort of a broad mix, ChemNet Base is great for organic compounds, novel, really strong in engineering, the Net Library or the NIST information, really good for thermophysical properties. Not all of them may have exactly what you're looking for, so it's good to have some options. Some of these databases are a bit more specific, structural information, different kinds of preparation strategies, engineering literature. And I'm looking at the clock and we're coming right up to the end of class, so I want to give people a chance to maybe ask a question if you have something that I either covered quickly or something else you'd like a little bit more detail on. Basically what I did is open the door for you. We got a lot of stuff available in the library. It's hard to get your head wrapped around any one of these just on them in the beginning. They'll be able to answer small questions for you and if you use them, say for the assignments, just remember that they exist because when you get bigger questions, it's nice to be able to go back and see how some of these might fit into part of your question or others could answer the bigger things that you're looking for, but they're really meant to do really big work and the little small things too. So as students, it's a chance to sort of experiment without too many consequences because the big news is when you're ready and you need them, they're here for you and I can certainly help you if you have specific questions going forward. Okay, so you certainly have an assignment. If you run into any kind of hiccups, please let me know. Those little slips of paper that you used, I'm going to use them tomorrow so you can hang on to them if you like. I'm just going to use them for tomorrow set. What you can do is use your individual password, login for SciFinder for doing searches going forward because these are good for today and tomorrow and then by five o'clock tomorrow, they retire. All right, well, thank you. Thank you for your time.