How to split an ebook pdf into separate chapter files
- Step 1Note each chapter's first and last page — Open the manuscript and record the absolute page numbers (physical position in the file, where page 1 is the first page) for each chapter's start and end. If front matter restarts the printed numbering, count from the physical first page.
- Step 2Drop the ebook into the splitter — Load the full manuscript into the PDF Split by Range tool. It parses in your browser and shows the page count and any embedded title above the options.
- Step 3Type the chapter ranges — Enter ranges separated by commas or semicolons:
1-22, 23-50, 51-78. Hyphen for start-end, bare number for a single page. Do not type chapter names — only digits, hyphens, commas and semicolons are parsed. - Step 4Run the split — One chapter PDF is created per token. Pages within each chapter stay in original ascending order. The result panel shows the chapter file count and total size.
- Step 5Download each chapter — Files download in order as
<book>.split-range.1.pdf,.2.pdf, and so on. Approve the browser's multiple-download prompt if it appears. - Step 6Rename and renumber as needed — Rename files to chapter titles in your file manager. If each chapter should restart its own pagination for an e-reader, run Page Numbers on each output.
Chapter-range parsing
The single range box is split on commas and semicolons; each token is one chapter file. Chapter labels are not parsed.
| You type | Output files | Result |
|---|---|---|
1-22, 23-50, 51-78 | 3 files | Ch1 (1–22), Ch2 (23–50), Ch3 (51–78) |
1-22; 23-50 | 2 files | Semicolon separates the same as a comma |
Ch1: 1-22 | 1 file | Ch1: is dropped; pages 1–22 extracted |
1, 2, 3-22 | 3 files | Title page, copyright page, then chapter body |
78-120 (78-page book) | 1 file | End clamped to 78 → just page(s) up to 78 |
What's preserved in each chapter file
Behaviour of pdf-lib copyPages on manuscript content.
| Element | Carried into chapter files? |
|---|---|
| Embedded fonts | Yes — copied with the page |
| Illustrations / cover art | Yes — at original resolution |
| Selectable text | Yes, if the source had a text layer |
| Page layout & margins | Yes — pages are not re-flowed |
| Bookmark / outline (clickable TOC) | No — chapter files have no outline |
| Document title metadata | Not carried into the new chapter files |
Tier ceilings for the source manuscript
Checked on the input PDF before splitting.
| Tier | Max file size | Max pages |
|---|---|---|
| Free | 2 MB | 50 pages |
| Pro | 50 MB | 500 pages |
| Pro + Media | 500 MB | 2,000 pages |
| Developer | 2 GB | 10,000 pages |
Cookbook
Chapter-splitting recipes for manuscripts. Page numbers are absolute file positions.
Three chapters for beta readers
Chapters at 1–22, 23–50, 51–78.
Range box: 1-22, 23-50, 51-78 Outputs: novel.split-range.1.pdf (Chapter 1) novel.split-range.2.pdf (Chapter 2) novel.split-range.3.pdf (Chapter 3)
Skip the front matter
Title, copyright, and dedication occupy pages 1–4; chapter 1 starts on page 5. Start the first range at 5 to leave front matter out of the chapter files.
Range box: 5-26, 27-54 Chapter files contain only chapter text (no front matter). To INCLUDE front matter in chapter 1 instead: 1-26, 27-54
Peel off the cover as its own file
The cover is page 1; you want it separate for an e-reader thumbnail.
Range box: 1, 2-23 Outputs: book.split-range.1.pdf (cover, 1 page) book.split-range.2.pdf (Chapter 1, pages 2-23)
Serialized release — one chapter at a time
Releasing chapter by chapter weekly. Run once for all, then publish files in sequence.
Range box: 1-22, 23-50, 51-78, 79-104 -> four chapter files, named .1 .2 .3 .4 publish one per week
Want a sample = first two chapters in ONE file
For a free sample you want chapters 1 and 2 combined into a single PDF, not two files. Use Extract Pages.
split-range '1-22, 23-50' -> TWO files For a single sample PDF: /pdf-tools/pdf-extract-pages '1-50' -> one combined sample
Edge cases and what actually happens
Chapter labels typed into the box
Labels ignoredCh1: or Chapter One is not a number and is dropped; only the numeric range is used. Output files are never named after chapter titles — rename them after download. Type ranges only.
Front matter included or excluded
By designWhether the cover, TOC, and copyright pages appear in a chapter file depends entirely on your range. Start the first range at 1 to keep them with chapter 1, or at the chapter's true first page to leave them out.
Clickable TOC missing from chapter files
Not preservedpdf-lib copies pages but not the bookmark/outline tree, so chapter files have no clickable table of contents. Any TOC pages inside a range are kept as pages; the navigation links are not.
Manuscript exceeds the tier page limit
Blocked (free 50 / pro 500)The whole book's page count is checked before splitting. A 300-page novel is blocked on Free's 50-page cap. Upgrade to Pro (500), Pro+Media (2,000), or Developer (10,000), or pre-cut with Split by Fixed Size.
Unpublished manuscript kept private
Local onlyAll processing happens in the browser with pdf-lib; the manuscript is never uploaded. This is important for embargoed or unpublished work. Only an anonymous usage counter is recorded if you're signed in.
Reversed or empty ranges
Dropped50-23 matches nothing and produces no file; blank tokens are skipped. If no valid token remains, the run fails with 'No valid page ranges parsed.' — verify your start comes before your end.
Reflowable EPUB-style behaviour expected
Not supportedA PDF has fixed pages; this tool splits by page, not by semantic chapter boundary or reflowable content. It cannot detect chapter breaks automatically — you supply the page numbers. For a true reflowable ebook, convert from a source format with dedicated EPUB software.
Source manuscript is encrypted
May fail to loadOwner-restricted manuscripts usually load (encryption ignored on read). A file needing an open password won't parse for page counting — remove it with Remove Password first, then split.
Frequently asked questions
Will the chapter PDF include the original cover page?
Only if you include it in a range. If the cover is page 1, use 1-22 to put it at the front of chapter 1, or 2-23 to leave it out and start at the chapter text. Or peel it off on its own with a single-page token: 1, 2-23.
Can I add the chapter name to the file?
Not in the tool — outputs are named <book>.split-range.1.pdf, .2.pdf, etc., in the order you typed the ranges. Rename them to chapter titles in your file manager. Don't type names into the range box; they're ignored.
What if my ebook has front matter like a table of contents?
Include those pages in your first range (e.g. 1-26 if front matter ends at page 4 and chapter 1 ends at 26), or skip them and start the first chapter at its true first page (e.g. 5-26). The clickable TOC links won't survive the split, but the TOC pages will if they're inside a range.
Does splitting reduce the quality of illustrations?
No. Pages are copied with pdf-lib's copyPages, not re-rendered, so illustrations, cover art, and embedded fonts are byte-for-byte identical to the manuscript. There's no recompression or downscaling.
Can the tool detect chapter breaks automatically?
No. It splits by the page ranges you provide, not by content. PDFs have fixed pages, so you must supply each chapter's start and end page. The advantage is precise control; the trade-off is you do the page-mapping.
How do I make a free sample of the first two chapters as one file?
split-range makes one file per token, so 1-22, 23-50 would give two files. For a single combined sample, use Extract Pages with 1-50, which keeps everything in one PDF.
Is my manuscript uploaded when I split it?
No. Everything runs in your browser with pdf-lib. An unpublished or embargoed manuscript never leaves your device. Only an anonymous usage counter is recorded if you're signed in.
How long can the ebook be?
Up to your tier's limit on the source file: Free 2 MB / 50 pages, Pro 50 MB / 500 pages, Pro+Media 500 MB / 2,000 pages, Developer 2 GB / 10,000 pages. A full-length novel typically needs Pro or higher.
How are out-of-range page numbers handled?
They're clamped to the document. On a 78-page book, 78-120 returns up to page 78 with no error; a start below 1 becomes 1. A reversed range like 50-23 matches nothing and yields no file.
Is the original manuscript changed?
Never. The tool always builds new chapter files and leaves the source untouched. Split the same manuscript as many times as you like with different range sets.
Can I split a chapter into equal-sized parts for serialization?
For equal-sized blocks, use Split by Fixed Size — set a pages-per-chunk count and it cuts the book into uniform pieces. Split by range is for the uneven, real-world chapter boundaries of a manuscript.
Why isn't the text selectable in my chapter files?
If the source manuscript was a scanned or flattened image with no text layer, the split chapters are images too. Run OCR on the source (or the chapter files) to add a selectable, searchable text layer.
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.