How to merge research papers into a single pdf reference
- Step 1Decide the bundle order — Choose how the bundle should read — by theme, by citation order, or chronologically. The merge panel has no drag-to-reorder, so the order you load files is the order they appear. Naming downloads with a prefix (e.g.
01_Smith2024.pdf) makes the sequence reliable. - Step 2Download each paper as a PDF — Save each article from the journal site, arXiv, your institutional repository, or a preprint server. Any valid PDF works regardless of publisher or source.
- Step 3Remove DRM/passwords from locked downloads if present — Some publisher PDFs are password-protected. If a file needs a password to open, it can't be merged directly — for papers you're licensed to use, strip protection with PDF Remove Password first, then merge.
- Step 4Drop the papers into the merger in order — Add files to PDF Merge in your chosen reading order. Each tile shows name, size, and page count. To fix a misplaced paper, click × to remove and re-add it.
- Step 5Run the merge and download the bundle — With at least two papers queued, click Process. The engine appends every page in order into one new PDF, all locally. Download the bundle (named from the first file, e.g.
Smith2024.merge.pdf). - Step 6Add pagination for citing within the bundle — Run the bundle through PDF Page Numbers so you can reference 'bundle p. 47' in notes — useful since the merge doesn't create a clickable table of contents.
What a merged reference bundle keeps — and what it doesn't
Based on the actual behaviour: pages are copied into a fresh document, so page-level scholarship survives but document-level navigation does not.
| Aspect | Behaviour on merge | Research implication |
|---|---|---|
| Equations, figures, typesetting | Preserved exactly — no re-rendering | Math and two-column layout look as published |
| Selectable text & embedded fonts | Preserved for text-based papers | You can still copy quotes and citations |
| Mixed page sizes | Preserved per page | Letter/A4/B5 sources coexist undistorted |
| Combined table of contents | Not generated | Use page numbers; build a TOC manually if needed |
| Each paper's own bookmarks/outline | Not carried over | No per-paper outline in the bundle |
| Hyperlinks & in-paper citations | Page content is copied as-is; internal links may not resolve across the merged file | Cross-paper links won't be created |
| Title / author metadata | Not copied — bundle starts clean | Set the bundle's own metadata if you want it |
Sources researchers merge from
Practical notes by source type.
| Source | Typical PDF | Merge note |
|---|---|---|
| Journal website (publisher) | Text-based, sometimes DRM/password | Unlock first if password-protected |
| arXiv / preprint server | Text-based, often Letter size | Merges cleanly; text stays selectable |
| Institutional repository | Text-based or scanned | Scanned theses won't be searchable unless OCR'd |
| Conference proceedings extract | May be A4/B5, figure-heavy | Size preserved; large files may exceed free cap |
| Scanned older paper | Image-only | Run PDF OCR for search |
Cookbook
Reference-bundle assemblies. Load order is reading order, so each example lists the exact sequence to drop files in.
Thematic literature-review bundle
Group papers by sub-topic so the bundle reads like a review. Prefix filenames to lock the order.
Load order: 01_Background_Smith2019.pdf 02_Background_Lee2020.pdf 03_Method_Patel2022.pdf 04_Results_Chen2023.pdf Merged: litreview.merge.pdf — themes in sequence.
Citation-order reading pack
Assemble papers in the order they appear in your manuscript's reference list for a focused re-read.
Load order = reference list order: Ref01_Anderson2018.pdf Ref02_Garcia2021.pdf Ref03_Wong2024.pdf Merged: refs.merge.pdf — read top to bottom.
Paginate the bundle for note-taking
The merge makes no table of contents, so add continuous page numbers to cite locations within the bundle.
Step 1: Merge → bundle.merge.pdf (62 pages)
Step 2: PDF Page Numbers, bottom-center, start 1
→ 1..62 across all papers
Now you can note 'see bundle p. 38' reliably.Make a scanned older paper searchable in the bundle
An archival PDF is image-only. OCR it so it's findable alongside the digital papers.
Step 1: PDF OCR on Historic_paper_scan.pdf, language English
→ searchable text layer added
Step 2: Merge with the text-based papers
The whole bundle is now searchable end to end.Pull one paper back out of the bundle
A co-author wants just the methods paper. Reverse the assembly by extracting its page range.
bundle.pdf — Patel2022 is pages 23-40 PDF Extract Pages → "23-40" → Patel2022.extract.pdf (18 pages) Or PDF Split by Range to break the bundle into per-paper files.
Edge cases and what actually happens
You expected a combined table of contents
Not generatedThe merge copies pages but does not build a navigational outline or TOC for the bundle, and it does not carry over each paper's own bookmarks. The result opens with no outline panel. Add continuous pagination with PDF Page Numbers for citing locations, and build a TOC manually in a desktop editor if your workflow needs one.
A publisher PDF is password-protected / DRM-locked
RejectedA paper that requires a password to open can't be read for merging and will fail to load. (Permission-only restrictions like copy-blocking are ignored, so those merge fine.) For papers you're licensed to use, remove the open password first with PDF Remove Password. Respect each publisher's licensing terms.
A scanned paper isn't searchable in the bundle
ExpectedMerging does not OCR, so a scanned (image-only) paper stays unsearchable even when bundled with text-based papers. Run PDF OCR on the scanned paper (before or after merging) to add a searchable text layer so the whole bundle is findable.
A figure-heavy paper exceeds the 2 MB free limit
Limit reachedArticles with many high-resolution figures often top 2 MB, which the free tier blocks. Reduce size with PDF Compress (Aggressive) (note this rasterises pages and loses selectable text — keep an un-compressed copy for quoting), or upgrade to Pro (50 MB / 500 pages per file).
More than 2 papers on free tier
Limit reachedFree merging combines 2 files per batch, once a day — a real literature review has far more. Merge in passes of two, spread across days, or upgrade to Pro for a single unlimited-file bundle (the practical choice for review-sized sets).
Papers ended up out of reading order
By designThere's no drag-to-reorder — papers merge in load order. Prefix downloads with a number, remove and re-add a misplaced file, or rearrange the merged bundle with PDF Reorder.
Internal hyperlinks/citations don't work across the bundle
ExpectedPages are copied as-is, but the merge doesn't rewire links to point across the combined file — an in-paper citation link generally won't resolve to a different paper in the bundle, and a paper's own internal links may not survive reliably. Use page numbers for navigation rather than relying on links.
Mixed Letter/A4/B5 page sizes in the bundle
ExpectedEach paper keeps its original page size, so size variation across publishers is normal and intended — nothing is scaled or cropped, preserving figure proportions. If you need a uniform size for printing the bundle, normalise with PDF Resize afterward.
Frequently asked questions
Will the table of contents from each paper be preserved?
No. Merging copies pages but not document outlines, so neither the individual papers' bookmarks nor a combined table of contents will appear in the bundle — it opens with no outline panel. For navigation, add continuous pagination with PDF Page Numbers; build a TOC manually in a desktop editor if you need a clickable one.
Will equations and figures lose quality when merged?
No. Each page is copied without re-rendering or recompression, so equations, vector figures, and two-column typesetting look exactly as published. Quality only changes if you deliberately run a paper through PDF Compress (Aggressive) to reduce file size.
Can I still select and copy text from the merged bundle?
Yes, for text-based papers (journal PDFs, arXiv preprints). Because pages are copied rather than rasterised, the text layer and embedded fonts are preserved, so you can copy quotes and citations. Only scanned (image-only) papers remain non-selectable — run PDF OCR on those to add a text layer.
Can I reorder the papers before merging?
Not in the merge panel — there's no drag-to-reorder. Papers combine in the order you add them, so load them in citation or reading order (prefixing filenames with 01_, 02_ helps). To fix an order mistake, remove and re-add the file, or rearrange the merged bundle with PDF Reorder.
Can I annotate the merged bundle afterwards?
Yes. The output is a standard PDF, so you can highlight and annotate it in any reader — Adobe Acrobat, Preview on macOS, Foxit, or your tablet's PDF app. Building one bundle first means your highlights and notes live in a single file rather than scattered across dozens.
How many papers can I merge at once?
Free merging combines 2 papers per batch, once per day, with each file up to 2 MB and 50 pages. A genuine literature review needs far more, so Pro — unlimited files at 50 MB / 500 pages each — is the practical tier. On free, you can also merge in stages or pre-compress large figure-heavy articles.
Can I merge a password-protected or DRM publisher PDF?
Only if it doesn't need a password to open. Permission-only restrictions (like copy-blocking) are ignored and merge fine, but a file requiring an open password can't be read. For papers you're licensed to use, remove the password first with PDF Remove Password — and always respect the publisher's licensing terms.
Are my unpublished manuscripts uploaded anywhere?
No. The merge runs entirely in your browser using pdf-lib; unpublished manuscripts, embargoed PDFs, and peer-review materials never leave your device. Only an anonymous run counter (no content) is recorded when you're signed in, and you can opt out.
Will hyperlinks and citation links work across the merged bundle?
Generally no. The merge copies page content but doesn't rewire links to span the combined file, so an in-paper citation link won't jump to a different paper in the bundle, and a paper's own internal links may not resolve reliably after merging. Use page numbers for navigation instead of relying on links.
A figure-heavy article is too large for the free limit — what can I do?
Articles with many high-resolution figures often exceed the 2 MB free per-file cap. Compress them first with PDF Compress (Aggressive) — but note it rasterises pages and removes selectable text, so keep an uncompressed copy for quoting. Pro raises the per-file limit to 50 MB if you'd rather preserve full quality and text.
Can I pull one paper back out of the bundle later?
Yes. Your original downloads are never modified, but if you only have the bundle, use PDF Extract Pages to pull a paper's page range into a new file, or PDF Split by Range to break the bundle into per-paper files.
Does the merged bundle inherit one paper's author and title?
No. The bundle is created fresh with empty metadata rather than copying any single paper's title or author — which avoids a misleading 'authored by' the first paper's author on a multi-source bundle. Set the bundle's own metadata afterward in your reader if you want it.
Privacy first
All PDF processing runs locally in your browser using PDF-lib and pdf.js. No file is ever uploaded — only metadata counters are saved for signed-in dashboard stats.