写科技论文，有基本功要求，主要是如何遣词造句，如何用标点符号，如何用缩略语等。关于这一方面的书有：The ACS Style Guide和The Elements of Style。可以说，这些都是科技写作的基本功，如果基本功不扎实，审稿人一眼就看出来了。但是，熟读这些书，把用法都背出来还是远远不够的。除此之外还有一些高级的方法和高级的指导书。曾经读到一些专门介绍科技写作的书，如Wiley出版社出版的The Arts of Scientific Writing，但是读了以后感到没有什么收获。后来认识到要学习到“高级”的写作方法，还是要回归到逻辑学、修辞学、市场营销学（广告学）、科学哲学上面来，从这些非化学的知识里面得到启发。
比如：写文章初稿的时候，一般人是按照自己的思路来写，想怎么写就怎么写。可是，高级的方法就是首先要分析很多问题：Is this research new? Are the data important and interesting? What is the point here? Since there are already a lot of papers, then why should we report our current work? Can the data sufficiently support the conclusions? Are the conclusions consistent with what was reported in the literature?这些问题有点象市场营销学：既然市场上已经有很多同类商品了，为什么我们要生产我们的商品？我们的商品和别人的商品有什么区别？怎么把“市场细分原理”运用到我们商品（文章）的销售中去？有这种critical thinking很重要！记得读博士的时候学习化学动力学，考试的形式之一是从文献里拿一篇文章来讲解。考官就会问：既然别人已经研究过这个体系，那么为什么这篇文章还要研究类似的东西？是它的结论推翻了以前的结论，还是它用了新的仪器方法，还是它的数据更加精确？如果只是数据更加精确，换汤不换药，那么意义在什么地方？
更重要的是，不是从自己的角度分析自己的文章如何如何好，而是从读者角度分析：Why should the reader read my paper? Why should people care about our research? How can our research benefit others' research? What can be done next by others based on our results? What questions will the reader have? If so, then how can we better write our paper to make the points clearer to the reader?这种思维方法如同准备做报告的思维方法一样：做报告之前，重要的不仅仅是想清楚自己要讲什么重要思想，而且还是要想清楚听众是哪些人？他们来听报告，希望得到什么样的信息？怎样讲才是对听众来说有价值的？怎样建立起自己的报告和听众的联系？有的人做报告就是说这个催化剂效果比那个好，听众睡着了。要提炼出更加有通用性的、方法学的东西，这样听众才会有兴趣。（关于这方面的方法学论述见哈佛商学院出版的Presentatios that Persuade and Motivate, 2004年出版。）这也和广告学、消费者心理学的原理类似，即要分析受众关心什么、需要什么。这个要点（从读者角度分析），说起来很简单，但是做起来却很难，因为这是一种很微妙的方法，一般人很难把握。
Don Dunbar写的What You Don't Know Can Keep You Out of College
p. 29. "For essays, showing your best self starts the same way as for interviews: You need to interest both yourself and your reader. But when you're writing, you don't have the reader sitting across from you, showing with his or her reactions whether you are getting through. For this reason, it's a good idea to make sure that the subject you're writing about is not just personal, but universal. Universal doesn't mean "very very big" and "important", it means that anyone--anyone in the universe, I suppose--can connect with it. It's universal the way type O blood is the "universal doner": anyone can receive it."
哈佛商学院出版的Face-to-Face Communicaions for Clarity and Impact
"Don't recite the facts--interpret them! People don't want a recitation. What they want to know is, What sense do you make out of this data? What is the conclusion? Do you have a logical flow in your thought process? And can we see that?"
哈佛商学院出版的Presentatios that Persuade and Motivate
"We humans are at once nobel and selfish. If you appeal soley to our self-interest, we will listen, and perhaps appreciate your words. But we won't respect you. To really get your audience on its feet--and rioting--you have to show them how self-interest and larger principles coincide, such that personal sacrifice is worth it if it becomes necessary."
"You need to find the way in which you and the audience are alike and make those clear early on. Your listeners will then be willing to open themselves to your message. It is way of building trust early on. Audiences want their speakers to have credibility, and they want to be able to trust them. You can't create the latter unless you find a way to conntect with your audience."
"In any communication task, you must understand what your audience needs to know, Your first step--long before you walk into the room--is to ask who the audience is and how they will use the information you provide. Then you can structure your presentation around those needs."
"The audience arrives wondering: Why should we care? Why is this important? So rather than jumping straight into the history of federal housing on Indian reservations, start by describing the current housing crisis. You get your listeners' minds working (How did this come about? How can we solve it?) in ways that give them context for the historical discussion.
Timothy W. Crusius和Carolyn E. Channell写的The Aims of Argument: A Brief Rhetoric
"The first step is to identify possible appeals to your readership. Keep in mind that good persuaders are able to empathize and sympathize with other people, building bridges of commonality and solidarity. To aid in your audience analysis, ask these questions:
-- Who are my readers? How do I define them in terms of age, economic and social class, gender, education, and so forth?
-- What type of attitudes or stances toward my topic do they have?
-- What in their background or daily experiences helps explain their point of view?
-- What are they likely to know my topic?
-- How might they be uninformed or informed about it?
-- How would they they like to see the problem, question, or issue resolved, answered, or handeled? Why?
-- In what larger framework--religious, political, economic--do they place my topic?"
在修改文章的时候，要走到纸头的“背面”去，用审稿人的眼光来挑剔地看自己的文章，给与自己的文章无情的轰击，指出这个不好，那个不清楚，还要做什么实验。带着疑问，去做实验，查文献，进行弥补漏洞。经过几次“反馈回路”，把文章的水平提高。相反，如果对自己的文章非常乐观和自信，有晕轮效应，得到稻草也以为是得到黄金了，那么就投不中。有的人自己做过很多回审稿人，有一种奇妙的感觉，知道投这个杂志应该达到什么样的水平。有的书上说，可以请懂行的人看一下文章，提出尖锐意见，然后进行修改。这里要说的是修改文章从审稿人的角度去修改，很有用处。可以说，很多文章投某个杂志，遭到退稿，其实按照审稿人的意见仔细修改以后，质量还是不错的，但是就是因为已经被退稿了，所以只能投其它杂志了。让我们想一想：假设这样的文章实现能得到懂行人的“模拟审稿意见”，把把脉，然后进行补做实验、精心修改，本来就能一次命中高档次杂志的！
Timothy W. Crusius和Carolyn E. Channell写的The Aims of Argument: A Brief Rhetoric
"Have a dialogue with yourself about your own writing.
1. Ask what you mean by the words that are central to the argument. Have you provided definations when they are needed?
2. Find the reasons, and note their relation to the thesis. Be able to state the connection, ideally, with the word "because": thesis because of reason.
3. Be able to state what assumptions lie behind your thesis and any of your reasons. Ask yourself, What else would someone have to believe to accept this as valid? If your audience is unlikely to share the assumption, then you must add an argument for it--or change your thesis.
4. Look at your comparisons and analogies. Are they persuasive?
5. Look at your evidence. Have you offered facts, expert opinion, illustrations, and so on? Have you presented these in a way that would not raise doubuts but eliminate them?
6. Consider your own bias. What do you stand to gain from advocating the position you take? Is your argument self-serving or truth-serving?
Because it is hard to be objective about your own work, getting a reading from a friend, classmate, teacher, or family member is a good way o see where revision would help. An unfocused reading, however, usually isn't critical enoughl casual readers may applaud the draft woo readily if they agree with the thesis and condemn it if they disagree. Therefore, have your readers use a revision checklist."
写好科技论文，除了满足基本功要求和用读者、审稿人的角度来反思自己文章以外，具体到写作方法来说，这里面也有很多“战法”。这正如拍电影、写小说、下围棋都是有一定方法的一样。关于段落的结构和写法可以见Sheridan Baker的The Practical Stylist。这本书讲到开头要从一个宽泛的事情或者概念收拢到本文具体到讲的东西。这本书还讲了每一段都要有主题句（topics sentence），每一段话讲一个意思，而每一段要有具体的功能！
Timothy W. Crusius和Carolyn E. Channell写的The Aims of Argument: A Brief Rhetoric
"Multiple paragraphs are generally required to develop and support a reason. The key thing to remember about paragraphs is that each one is a unit that performs some function in presenting the case. You ought to be able to say what the function of a given paragraph is--and your reader ought to be able to sense it. Doe it introduce a reason? Does it define a term? Does it support a reason by setting up an analogy? Does another paragraph support the same reason by offering examples or some hard data or an illustrative case?"
"Not all paragraphs need topic sentences to announce their main point. Worry instead about opening each paragraph with some hints that allow readers to recognize the function of the paragraph. For example, some transitional word or phrase could announce to readers that you are turning from one reason to a new one. When you introduce a new reason, be sure that readers can see how it relates to the thesis. Repeating a key word or offering a synonym for one of the words in the thesis is a good idea."
我们经常发现很多写得不好科研论文的段落没有功能，或者一段话有两个功能！现在我来举例说明段落的主题句和功能。22. H.F. Yin, Z. Ma, S.H. Overbury, S. Dai*, Promotion of Au(en)2Cl3-Derived Au/Fumed SiO2 by Treatment with KMnO4, Journal of Physical Chemistry C, in press.
http://dx.doi.org/10.1021/jp800797t
这篇文章Introduction的第三段写得很基本，是白话文，但是里面有起承转合在里面：第一句是总起句。接下去，第二句和第三句的功能是举例说明总起句总结的内容：第二句指出在文献中，别人做了什么。这句话用了一个"For one..."，表示这句话的功能是举例；第三句指出我们小组和别人以前还报道了什么。既然别人和我们小组做了这个，做了那个，那么为什么发表本文有必要呢？第四句笔锋一转，说："Nevertheless, even though..."其功能是指出文献（或者说“市场”）中缺乏的东西（或者说“产品”），为第五句埋下伏笔。第五句"Hence, the promotional effect based on a better starting point (i.e., highly active Au/SiO2 synthesized by advanced methods16-22) may be further considered, and new methodologies for the installation of appropriate promoters may be developed."指出根据以上逻辑分析，在“市场”上，什么样的“产品”是需要的。这句话的好处之一是用了"hence"(therefore)显示逻辑关系。好处之二是它“预测”了市场上什么样的东西是需要的。这当然不能乱写，胡乱预测，因为预测的东西，通过做实验，在本文中得到了实现。如果作者不准备报道什么东西，就不要在Introduction里面胡乱预测。也就是说，预测的东西在后面是要有“呼应”的，如果没有呼应，审稿人就会说他被欺骗了。
"Attempts have been made to modify SiO2 supports before loading gold.28-35 For one, Nieuwenhuys and coworkers reported that the T50 (reaction temperature at which 50% of CO molecules are converted to CO2) value of Au/SiO2 in CO oxidation is 240oC, but when SiO2 is modified by CoOx, LaOx, or CeOx and gold is loaded thereafter, the T50 values decrease to 185, 135, and 115oC, respectively, indicating the promotional effect of these additives.28 Our group29,31 and others28,30,32-35 developed Au/TiO2/SiO229-33 and Au/CoOx/SiO234,35 catalysts for CO oxidation. Nevertheless, even though the addition of certain promoters may improve the activity, the promoted catalysts are still not particularly active due to the unadvantageous starting point: the deposition-precipitation method does not work well for making active Au/SiO2 catalysts.29 Hence, the promotional effect based on a better starting point (i.e., highly active Au/SiO2 synthesized by advanced methods16-22) may be further considered, and new methodologies for the installation of appropriate promoters may be developed."