Sharing and hosting policy FAQ

We have recently updated our posting policy and introduced a new hosting policy. We announced this change on our community and information site, Elsevier Connect, and have provided some FAQs below based on feedback from our customers, researchers and the wider academic community. If you have further questions or feedback, please contact us via or see our policy handout and our "what's changed" handout.

Sharing Policy

We have formally supported the STM Article Sharing Principles, alongside other publishers, and are keen to provide additional clarity for researchers, libraries and also to other hosting platforms about how to share and aggregate research published by Elsevier. We have started by updating our posting policy and introducing a new hosting policy and are focused on addressing some of the shared challenges the industry faces such as developing the technology to facilitate sharing across different platforms.

We recognize that authors want to share and promote their work and increasingly need to comply with their funding body and institution's open access policies. We wanted to update our policies to provide clearer and more consistent guidelines that support these needs and are aligned with the rest of the publishing industry.

We are broadening the ways researchers can share their research; we continue to support sharing of preprints, accepted manuscripts, and final publications and provide simple guidelines for authors about how they can share at each stage of their workflow. More specifically:

- We have removed the need for non-commercial hosting platforms to enter into agreements with us.

- We are also providing a range of options for researchers to share their work publicly, including a Share Links service which provides 50 days free access to the final article on ScienceDirect.

- We are also providing a wider range of ways for researchers to share their work privately during the journal’s embargo period, such as in their Institutional Repository's private workgroups and on sites such as Mendeley.

See our handout "what has changed" (PDF).

Green open access is the result of making a version of a subscription article available to everyone. The sharing policy gives authors guidelines about how they can do this and also how to share their articles in other ways. For example, they may want to share their latest paper with colleagues, students, or other members of an online working group.

Every author who publishes with us, retains important rights including the right to use their paper for personal use, or for internal use at their institution. The sharing policy is aligned with these author rights and provides additional ways authors can share their research, for example on their institutional repository for public access after embargo.

Yes. We are already listed as a "green" publisher and we don't expect this to change with the new updated policy announcement.

You can find the embargo period information for all Elsevier journals on the journal homepage or alternatively download a complete list here: The embargo period begins from the date the article is formally published online in its final and fully citable form (i.e. online publication date).

Elsevier has a number of funding body agreements in place to help authors publishing in Elsevier journals to comply with their funder or institutional open access policies. Please click here for more details on specific agreements.

It was not our intention that institutional repositories and other non-commercial repositories take retrospective action, and we will not enforce compliance for older content on these sites. We want to help non-commercial sites implement going forward, and will be providing tools and services to help you with this – for example tagged manuscripts. If there are implementation elements which you find challenging, we would like to help and encourage you to contact us via

Preprint: Author's own write-up of research results and analysis that has not been peer reviewed, nor had any other value added to it by a publisher (such as formatting, copy-editing, technical enhancements, and the like).

Accepted manuscript: the version of an article that has been accepted for publication and which typically includes author-incorporated changes suggested during submission, peer review, and editor-author communications. They do not include other publisher value-added contributions such as copy-editing, formatting, technical enhancements and (if relevant) pagination.

Published journal article: This is the definitive final record of published research that appears or will appear in the journal and embodies all value-adding publisher activities including peer review co-ordination, copy-editing, formatting, (if relevant) pagination, and online enrichment.

Private sharing is about the audience, for example sharing with a colleague or with an invitation-only online group. Mendeley, for example, is one platform that provides support for both private group sharing and public sharing. Click here for more details.

Our policy has an added requirement to apply a Creative Commons non-commercial license to publicly posted accepted manuscripts. This ensures readers understand how they can reuse the version of article they are accessing. We have provided some easy guidance on how to attach a license here.

Yes. Corporate researchers may share articles they have written in line with the sharing policy. Corporate researchers should check the terms of their organization’s license agreements for guidance on how to share other articles access via ScienceDirect (and other Elsevier platforms).

Actually, embargo periods have been used by us – and other publishers – for a very long time and are not new. We’ve had these in place for all our journals since 2004, and they have not been affected by the policy change. What has changed is that our former policy required Institutional Repositories to have an agreement with us, which included the embargo provisions. Our new policy no longer requires an agreement with Institutional Repositories, provided the embargo periods are used.

Yes, but given the growth of green open access it is very rarely the case that no funder or institutional policy applies, and in our experience many repositories in this category respected our embargo periods anyway. We continue to permit immediate self-archiving of the accepted manuscript in an author’s institutional repository which may be shared for internal institutional uses or as part of an invitation-only research collaboration work-group and public access after the embargo period. This is now true for all institutional repositories, not only those with which we have agreements or those that do not have mandates. Our new policy is designed to be consistent and fair for everybody, and we believe it now reflects how the institutional repository landscape has evolved in the last 10+ years.

Our journal specific embargo periods are evidence-based and typically range from 12-24 months, however there are exceptions which can be both longer and shorter than 12/24 months. Ideally, embargo periods should be set on a title-by-title basis by publishers. However, we recognize that other stakeholders – in particular funders – seek influence over embargo lengths. We have worked in partnership with funding bodies for many years on open access, and are committed to continued collaboration. We are aware of the many new funding body policies that have emerged in the last year with 12 month embargo periods and will factor this in during the 2015 review.

No, preprints can be used anytime anywhere by authors. We encourage authors seeking to publish in Cell Press, The Lancet, and some society owned titles to check the author section on the journal homepage for additional information.

To submit a manuscript in arXiv, the author can either grant ArXiV the non exclusive right to distribute the article or use a CC-BY or CC-BY-NC-SA license ( Since with CC-BY-NC-ND user license, authors can grant ArXiv the right to distribute the paper, accepted articles from Elsevier’s journals can be posted on arXiv within their policy.

No. We have removed the need for an institution to have an agreement with us before any systematic posting can take place in its institutional repository. Authors may share accepted manuscripts immediately on their personal websites and blogs, and they can all immediately self-archive in their institutional repository too. We have added a new permission for repositories to use these accepted manuscripts immediately for internal use and to support private sharing, and after an embargo period passes then manuscripts can be shared publicly as well.

The key snippet from the agreements for articles published under the subscription model is:

"Retention of Rights for Scholarly Purposes"
I understand that I retain or am hereby granted (without the need to obtain further permission) the Retained Rights (see description below), and that no rights in patents, trademarks or other intellectual property rights are transferred to. The Retained Rights include the right to use the Preprint, Accepted Manuscript and the Published Journal Article for Personal Use, Internal Institutional Use and for Scholarly Sharing. In the case of the Accepted Manuscript and the Published Journal Article the Retained Rights exclude Commercial Use (unless expressly agreed in writing by ), other than use by the author in a subsequent compilation of the author’s works or to extend the Article to book length form or re-use by the author of portions or excerpts in other works (with full acknowledgment of the original publication of the Article).
Click here to view a sample agreement (PDF)

Authors who choose to publish gold open access with Elsevier are offered their choice of 2 CC licenses. We don’t offer this choice on self-archived subscription manuscripts and instead use a non-commercial CC-BY-NC-ND license. Green open access must work in harmony with the subscription business model and commercial subscriptions are important for many journals, so the use of a non-commercial CC license is in place as an important safeguard. Authors when asked in surveys often choose NC ND of their own volition (see the 2014 Taylor and Francis study).

Through talking to and learning from researchers we believe we have a policy that is flexible and comprehensive and crucially strikes a balance between supporting authors to share their research in myriad ways, whilst ensuring journals can continue to be sustainable. We are experienced in understanding the impact of particular types of article sharing and publically available evidence demonstrates that certain types of sharing do impact on the sustainability of the journals that authors choose to publish in. This is particularly important for green open access where we are finding a balance to support large scale sharing of subscription content for free.

The sorts of evidence available to factor into embargo decisions include:

1. Usage Evidence: In 2014, Phil Davis published a study commissioned by the Association of American Publishers which demonstrates that journal article usage varies widely within and across disciplines, and that only 3% of of journals have half-lives of 12 months or less. Health sciences articles have the shortest median half-life of the journals analyzed, but still more than 50% of health science journals have usage half-lives longer than 24 months. In fields with the longest usage half-lives, including mathematics and the humanities, more than 50% of the journals have usage half-lives longer than 48 months. Read more

2. Evidence for the link between embargo periods, usage and cancellations:

A 2012 study by the Publishing Research Consortium explored the impact of a potential global 6 month embargo period. The survey received a 26% response rate. Librarians reported that “usage statistics are often used to help determine whether or not to cancel a journal, but most respondents said that they are usually appraised in tandem with other variables.” The conclusion drawn was “that STM publishers would fare better than AHSS publishers. Overall, STM publishers could expect to retain full subscriptions from 56% of libraries; AHSS publishers could expect to retain full subscriptions from 35% of libraries. STM publishers could expect 10% of libraries to cancel subscriptions altogether, and AHSS publishers could expect 23% of libraries to cancel subscriptions altogether. STM publishers could expect reduced (or no) revenues from the remaining 34% of libraries; AHSS publishers could expect reduced (or no) revenues from the remaining 42% of libraries. Most publishers would be obliged to review their portfolios; and a substantial body of journals, especially in AHSS subjects, would cease or be financially imperiled.”

A 2012 study by ALPSP was a simple one-question survey: "If the (majority of) content of research journals was freely available within 6 months of publication, would you continue to subscribe?" The results “indicate that only 56% of those subscribing to journals in the STM field would definitely continue to subscribe. In AHSS, this drops to just 35%. This 2012 study builds on earlier, more nuanced, studies undertaken for ALPSP in 2009 and 2006 and by PRC in 2006. The 2009 ALPSP study (see the next to last bullet) found that "overall usage" is the prime factor that librarians use in making cancellation decisions. The 2006 ALPSP study (see points 7 and 8) found that "the length of any embargo" would be the most important factor in making cancellation decisions.

A 2006 PRC study (see pages 1-3) shows that a significant number of librarians are likely to substitute green OA materials for subscribed resources, given certain levels of reliability, peer review and currency of the information available. With a 24 month embargo, 50% of librarians would use the green OA material over paying for subscriptions, and 70% would use the green OA material if it is available after 6 months.

The evidence linking journal usage patterns to library cancellation patterns is quite well documented, for example:

ENSSLE, HR; WILDE, ML. So you have to cancel journals? statistics that help. Library Collections, Acquisitions, & Technical Services. 26, 3, 259, Sept. 2002.

NIXON, JM. A Reprise, Or Round Three: Using a Database Management Program as a Decision-Support System for the Cancellation of Serials. Serials Librarian. 59, 3-4, 302-312, Oct. 2010.

GALLAGHER, J; BAUER, K; DOLLAR, D: Evidence-based Librarianship: Utilizing Data from All Available Sources to Make Judicious Print Cancellation Decisions. Library Collections, Acquisitions, and Technical Services. 29, 2, 169-179, 2005.

GREEN, PR. Monitoring the Usage of Science and Engineering Journals at the Edward Boyle Library, University of Leeds. Serials Librarian, 25, 1-2,
169-180 (1994).

While research libraries do very often have several big deal packages that take lots of individual journals off the table at renewal/cancellation time, it¹s also true that most research libraries still maintain hundreds if not thousands of individual journal subscriptions and have to make renewal/cancellation decisions about them individually. According to Rick Anderson, Dean of Libraries at the University of Utah, on the liblicense list “When we have to make a cut, the first thing we ask is  "What's the cost per download, and how does it compare to the cost of ILL or document delivery?".

3.Experiences of other journals: For example the Journal of Clinical Investigation which went open access with a 0 month embargo in 1996 and lost c. 40% of institutional subscriptions over time. The journal was forced to return to the subscription model in 2009. Other examples include the Annals of Mathematics, the Journal of Dental Research, the American Journal of Pathology, and Genetics.

The Creative Commons license allows users to reuse your manuscript for non-commercial purposes. Full details about this license can be found from the Creative Commons website. In general, the license permits users to read, print and download the manuscript but also have this redistributed or displayed, for example in a repository. Users can also text and data mine your article and also translate it and reuse portions or extracts in other works. However, reuse is only for non-commercial purposes and as such posting on commercial websites, or selling the manuscript would not be allowed.

It is also important to make sure that readers and users can find and cite your final version of your article from ScienceDirect. The way to do this is to include the Digital Object Identifier (DOI) link in your posted article. A DOI is a standardized method for identifying an electronic object and you can easily find your DOI under the title of your article. To convert a DOI to a Web address, add the following URL to the DOI: followed by your DOI number. We recommend you include this information to your title page or header/footer.

No, Elsevier does not view publication as an academic thesis as prior publication. Please note that Cell Press, The Lancet, and some society-owned titles have different policies on prior publication.  For further information on Elsevier’s prior publication policy please see: Policies and Ethics for Journal Authors (Multiple, redundant or concurrent publication).

Elsevier supports sharing of articles in private workgroups in the following applications: Zotero, Mendeley, Refme, CiteUlike, Sparrho and Jabref.

Hosting Policy

We believe we should be working with other platforms which host content to build best practices and industry standards which enable researchers to share in simple and seamless ways. We have supported the STM Article Sharing Principles and have now updated our policies by introducing a new hosting policy, with the aim to give clear guidelines for sites who wish to aggregate and make available content self-archived by researchers.

Our hosting policy includes the following in the way commercial use or posting of articles is defined:

- For commercial gain. For example by associating advertising with the full-text of the article, by providing hosting services to other repositories or to other organizations (including where an otherwise non-commercial site or repository provides a service to other organizations or agencies), or charging fees for document delivery or access

- To substitute for the services provided directly by the journal. For example article aggregation, systematic distribution via e-mail lists or share buttons, posting, indexing, or linking by commercial companies for use by customers of such companies (e.g. pharmaceutical companies and physician-prescribers).

Our new hosting guidelines make it clear that IRs can host their researcher's:

  • Metadata and links
  • Preprints
  • Accepted manuscripts immediately for internal use and for private sharing.
  • Publicly make hosted accepted manuscripts available after embargo and with a CC-BY-NC-ND license
  • Gold open access articles.
  • Subscription articles for private sharing as per their ScienceDirect agreement

We also have a number of hosting services available for institutional repositories. Please click here for more information.

No. This is one of the ways in which our new policy is more flexible for all repositories than our old policy was.

We have been working with a number of Institutional Repository Managers to better understand their needs and to provide services which will make it easier for them to implement detailed aspects of the hosting policy.We are piloting automated ways for repositories to easily host and link to content. See our Institutional Repositories page for more information.

This means:

- Use by the author's institution for classroom teaching at the institution and for internal training purposes (including distribution of copies, paper or electronic, and use in coursepacks and courseware programs, but not in Massive Open Online Courses)

- Inclusion of the Article in applications for grant funding

- For authors employed by companies, the use by that company for internal training purposes

Millions of researchers already have access to ScienceDirect. We have an API which has this information and we are willing to share this with hosting platforms so that they can display the best available version to their users. The process is simple, the platform would need to include the Crossref DOI or the Elsevier article identifier, called PII in the API request. For non-commercial platforms please email us via

The ScienceDirect APIs are free for everyone to use. To register your interest, please click here.

No, this is entirely incorrect. The two approaches are entirely complementary. Some of our IR partners have chosen not to host accepted manuscripts, but this is their choice.


We do regularly scan and monitor posted content and we may formally request platforms to take down incorrectly shared content. We are hopeful that the new policies will reduce the need to send these notices and have developed a wide range of tools and technology to make sharing on platform such as Institutional Repositories easier. We also support the STM Article Sharing Principles to enable private scholarly sharing on commercial platforms.

No. Our aim is to work with platforms rather than authors and we have never sent notices directly to authors. We may from time to time contact individual researchers to let them know that we have instructed the hosting platform to remove an incorrectly posted article and to let them know about alternative services.