Guide on how to write a scientific article for publication in Engineering Fracture Mechanics

1. Foreword

Manuscripts submitted to Engineering Fracture Mechanics are expected to conform to certain minimum standards before they can be considered for publication. Articles which do not conform to these minimum standards will usually be rejected by the editors without even being sent for scientific peer review. These guidelines are provided for those authors who have little or no experience in preparing scientific papers to ensure that their submission satisfies these minimum standards. Engineering Fracture Mechanics contains excellent examples of well written papers which should serve as models for the not-so-experienced writer.

Before setting forth these minimum standards, we note two important imperatives. First, a manuscript should never be submitted for publication without having passed the author's own internal quality assurance mechanism. This may amount to a check by the author's supervisor(s) or manager, or may be a more rigorous procedure. Furthermore, the paper should be read thoroughly by all authors (not just the first author) as all authors are jointly responsible for the contents of the paper.

Second, before writing a manuscript to be submitted for publication in an international journal, you should ask yourself whether the purpose in writing the article is merely to extend your list of publications for your next evaluation or for your CV. If that is the case, then stop reading here. If it is your firm belief that your work may contribute something relevant to the science of your field, then you may continue.

2. Some General Comments

Please do not just sit down and describe what you have done. Sometimes, even well-written manuscripts are rejected by a reviewer due to the lack of novelty, i.e. the reviewer asks: What did I learn from this paper that I did not know before?

Many reviewers will notice when an author has recently published a very similar paper which is only marginally different from that he/she has been asked to review. This attempt to oversell what has been done is regarded as bad practice. Prospective authors should note that, in a group of 3 or 4 reviewers, there is a high likelihood of such a well informed reviewer.


  • Never submit unchanged a paper that has already been rejected by another journal it may land on the desk of a previous reviewer who had rejected your manuscript.
  • Never submit a manuscript to two journals in parallel, hoping that it will be accepted by one. Apart from the fact that this is grossly unfair (you waste the time of the referees of one journal) you may also hurt yourself we know of cases where both submissions were sent to the same referee for review! Some journal editors may refuse to consider future submissions from authors who deliberately ignore this rule.

Write in a clear and straight style - obscure formulations and overly long phrases will quickly annoy the reader, and may lead the reader to believe that the author's thoughts lack clarity. Sometimes some self-criticism may help; you should ask yourself: Do I understand what I have just written? Reading papers from other authors may help develop a good writing style. (Of course, one should not copy text from the work of another author.)

A publication should not be too long, even when a journal does not prescribe a maximum length. Nowadays, readers usually do not have the time to read lengthy papers: convey just one key message with a concise text. Exception: A review paper needs much more room than an original communication.

Recent statements on the ethics of scientific publications require that no advertising material shall be included. Please observe the rules of ethics for a scientific publication provided in the journal's Guide for Authors.

For good reasons, a scientific publication should have a certain structure which will be described in the following sections.

3. Title

The title should be as short as possible and be descriptive of the content of the paper. Normally, the title should not contain acronyms, since the vast majority are not known by the general readership. This holds in particular for those acronyms invented by the author.

4. Abstract

The abstract is a concise overview of the content of the paper, and explains the key problem being addressed, the approach to this problem, and what has been achieved. The abstract should also not contain acronyms, nor should they be used in the section titles. References to other publications should be avoided in the abstract; their place is in the main text. An exception to this rule may be reference to very important work by another author which was used as the basis for the present paper, e.g. '...R.A. Smith s method of...'

5. Keywords

The keywords should cover the content of the paper. Their importance is frequently underestimated: The keywords provide the profile of your paper for data banks, and should therefore be selected with care. Engineering Fracture Mechanics has a long list of keywords from which those for your paper should be selected.

6. Nomenclature (List of Symbols)

A publication containing numerous formulae and hence mathematical symbols requires a list of these symbols. Otherwise, a reader is always in danger of getting lost. This list MUST be placed on the second page of your manuscript and, if acronyms are used, these should be provided in a separate section of the nomenclature.

7. Introduction

Essentially, this section provides the justification of the work you have done. The existence of your paper implies that there is an unsolved or even a new scientific problem to which you have dedicated your work. From this it follows that you have to provide a concise presentation of the state of the art of the field, citing the relevant literature. But, do not only provide a long list of references. It should become clear in this section that you have digested what you have referenced.

Warning: It appears, particularly in papers from younger authors, that literature older than e.g. five years is disregarded. As a consequence, older work is overlooked which may already have solved your problem. This is an important issue: We quite often receive papers 're-inventing the wheel'. This and the issue of non-digested work of others is a point where reviewers may be particularly fussy, perhaps because you have failed to reference some of their work.

At the end of this section, you should describe briefly the gap you have filled and how you have done it.

The following section contains guidelines for papers based primarily on experimental work; additional guidance for papers based primarily on theoretical and computational work will be given later in the following section.

8. How to Describe Your Experimental Program

Here (or alternatively at the end of the Introduction) the presentation of the goal of the work and the strategy used to achieve it is described.

8.1 Material

Describe why you chose this material (or materials) and why you chose the particular methods used to characterise the material(s) in the present work.

8.2 Specimens (Test Pieces)

It may be necessary to show drawings of the specimens with units (SI units only). If standard test specimens have been tested, then reference to the relevant standard may suffice. For a very large test program, a matrix-like table may provide a good overview. Specimens may be taken from ingots, semi-finished products, or components; their orientation and location in the source material should be specified. ASTM and ISO provide standard designations for these.

8.3 Experimental Procedure

The following information is required:

  • The kind of tests that were done and the test conditions, for example
    • Test temperature
    • Loading rate
    • Environment
  • The parameters that were varied, the quantities that were measured and their measurement methods with accuracy, precision, resolution etc., and the quantities that were evaluated and the procedures to evaluate them must also be described
It is of paramount importance that the whole test procedure and/or additional theoretical work are presented such that it is possible for someone (with the necessary expertise) to duplicate your tests and/or theoretical work. This is an important contribution to good scientific practice.

8.4 Results

The best way to present the results is to show them in the form of clear diagrams. Usually, results should not be presented in diagrams plus tables. Sometimes, however, it may be useful to have the results available in the form of numbers, in particular, if they will be used for further evaluations. Such tables may be placed in an appendix. In any case, double presentation should be considered with care.

Of course, the experimental results should also be described in words, however, not in the sense: 'The curve for the first test series increases up to and then decreases...' etc. Don't write what the diagrams already describe clearly.

Additional Guidance for Theoretical/Computational Work

The above comments and recommendations also hold for theoretical and computational work. In papers based on computational work, usually the finite element method, the mesh (or net), the FE element type used, as well as the boundary conditions and input parameters should be provided. Someone experienced in numerical analysis should know about the limitations of his/her methods and models. Presenting numerical 'accuracies' with an unsupportable number of significant figures demonstrates lack of understanding and knowledge of limitations of the computational methods employed.

In papers on analytical work, you should not just present derivations over a number of equations and pages; explanatory text illustrating the course of your thoughts is absolutely necessary; otherwise you will lose the attention of your reader. The reader should be able to understand the content of your work without following the derivations in detail. The correctness of equations has to be checked by intermediate calculations.

As in the case of experimental work, simply describing some numerical or analytical derivations without considering the theoretical (physical) background is not normally sufficient to justify a publication. Simply reporting pages/tables of numerical results, like endless data from experimental work, with no effort to determine, or at least hypothesize the underlying meaning, or to fit empirical equations to the results, demeans your work and is worthless to the reader.

Comparing your numerical result to someone else's numerical result may be informative, but it PROVES nothing. Even generally accepted numerical results have subsequently been shown to be incorrect (you may have made the same mistake or are using the same bad assumption as the original authors). Verification by means of comparison to already well-known solutions and validation by means of comparison with experiments, is mandatory.

Good experimental work, even when it uses well established methods, will at least deal with new subjects tested (materials, components, etc.). For analytical work, where this is not the case, originality of theoretical contributions may be a critical issue.

9. Discussion

This section requires much skill as it brings order and interpretation to the results. Studying the effects of the variation of the parameters on the experimental results examined leads to conclusions about the mechanisms of the events studied, and to the development of physical models. This in turn may serve as the basis for further analytical and/or numerical work, either in the paper presented or in future work. This is certainly an important aspect of your contribution: the mere reporting of experimental results without an attempt to search for the underlying mechanisms is of little value and such work will not normally be accepted. It should be borne in mind that every experiment is based on a model, even if you are not aware of it.

The purpose of this section is to demonstrate the gain in knowledge from your work, and to put your results into perspective by comparing them to the state-of-the-art described in the Introduction. Your results will thereby contribute to the overall knowledge of the area under investigation.

Whenever possible, schematic representations of the model(s) developed are a very versatile tool for conveying the message of the work undertaken. These can be real eye catchers, attracting the reader s attention and motivating him/her to read your article.

A compilation of numerous diagrams and colour pictures does not make a scientific result. A journal article is not a PowerPoint presentation. Ordering the data and systematic presentation and interpretation of the results is an intellectual effort that has to be performed by the author and not by the reader.

There are cases where it may be preferable to discuss the relevance of the results in conjunction with their presentation in the section on results.

10. Conclusions

As its title implies, the conclusions drawn from the research work described in the paper are described in this section. Please be concise. A frequently used format is to start with a few phrases to summarise the work program, and then to list the main conclusions with bullet points. Sometimes this section is misunderstood to be a repetition or restatement of the Abstract. This is not acceptable practice. A portion of the Abstract is a summary of the Conclusions section.

11. Acknowledgement(s)

Acknowledgements are due to financial sponsors as well as colleagues who supported the author in one way or another, below the level of a co-authorship. A 'Thank you' in print is an inexpensive way of expressing professional collegiality.

12. References

Before you compose your list of references (and also before you cite them in the main text) you should check the reference style required by Engineering Fracture Mechanics. This information is provided in the Guide for Authors at:

13. Authors Who Are Not Native English Speakers

This is a somewhat delicate point: On the one hand, nobody expects authors who are not native English speakers to provide a manuscript in perfect English. On the other hand, a manuscript in below average language gives a poor impression, and the reader might conclude that the quality of the content follows the quality of the language. If the scientific content of a manuscript can not be sufficiently evaluated because of the poor quality of English, the manuscript will be rejected.

In any case, for the final acceptance of a manuscript, sufficiently correct English is an indispensable necessity.

Frequent mistakes are the improper use of the article ('a' and 'the' ) and misuse of tenses (for example switching from past to present tense in the course of a sentence). In many cases, even seemingly small deviations from the correct language may obscure the actual meaning of a statement. If you are not experienced in writing English manuscripts, you may consult the Instructions for Authors in the journal.

14. Concluding Remarks

Summarising the above statements: When you write a paper, put yourself into the position of the reader, so that you can explain what you have done:

  • Why did I deal with this subject?
  • What did I do?
  • What did I obtain from my work?
  • How are the results to be interpreted?
  • Sloppy writing, e.g. numerous typos, poor style, tiny illustrations, poorly written equations;
  • Lengthy text containing redundant statements;
  • Too many similar illustrations; and
  • Sloppy appearance of the manuscript.
We have given you some guidance on how to organise your manuscript. However, the science is your job!

Editors-in-Chief and Editorial Board of Engineering Fracture Mechanics