Free Patents Online: IP Research and Community

Search Tutorial

Quick Search

Quick search lists the most common search fields. You can search for a word or a simple phrase or patent number in any of the listed fields. You can search in more than one field at a time. You can also use wildcards to enhance your searches in quick search (see wildcards for tips and examples for searching with wildcards). Boolean operators and nesting are not allowed in quick search. Quick search also has additional features like word stemming, sort order and date range to further enhance your search. In both quick search and expert search, the collection searched is defaulted to US patents. To change the collection searched, just select the required collection from the collection options provided.

Expert Search

Expert search helps you create complex queries to search multiple fields using field abbreviations. You can also include boolean operators, nesting and wildcards in your expert search query. To add value to your searches, FreePatentsOnline incorporates proximity searching and search term weighting features in expert search. Word stemming, sort order and date range options are available to further enhance your search. In both quick search and expert search, the collection searched is defaulted to US patents. To change the collection searched, just select the required collection from the collection options provided.

Sort Order

You can select to sort the documents in your results list either chronologically or according to their relevancy. If you select chronological sort order, documents in the results list would be arranged chronologically, with most recently published documents on the top of the list. You can also select to sort the documents in the results list according to their relevancy. Relevancy sorting is based on a "relevancy score" assigned to each document. Relevancy score is based on frequency and positioning of keyword(s) in the document. Relevancy score ranges from 0 to 1000. Documents with higher relevancy score would be placed at the top of the list.

Boolean Operators

Boolean operators help you combine words or phrases to broaden or narrow your search. You can use boolean operators to specify what you want and what you do not want while searching. FreePatentsOnline allow AND, OR and NOT operators.


Boolean Operator Example Function
AND Mobile and Telephone Finds documents which have both mobile and telephone.
OR Mobile or Telephone Finds documents which have either mobile or telephone or both.
NOT Mobile not Telephone Finds documents which have only mobile. The NOT operator excludes telephone completely.


Order of precedence for boolean operators is:
1. NOT
2. AND
3. OR
Which means that in a query with NOT, AND and OR, terms connected by NOT would be processed first, followed by terms connected by AND and lastly, those connected with OR. Example: auto or manual and gear - Since AND is processed before OR, the query would be interpreted as auto or (manual and gear). Note: FreePatentsOnline search engine is not case sensitive for boolean operators.

Nested Queries/Creating queries with Parentheses or Brackets

You can enclose search terms and their operators in parentheses or brackets to specify the order in which they should be processed. Information within parentheses is read first and then information outside parentheses. If there are more than one set of parentheses, the innermost set of parentheses would be read first, then the next and so on until the whole query is interpreted. The order of precedence for parenthesis and boolean operators is:
1. Parentheses
2. NOT
3. AND
4. OR

Nested query How it is interpreted (following order of precedence)
Hair and dye or colorant (Hair and dye) or colorant
(one or two and layer) and insulation (one or (two and layer)) and insulation

[Back to top  Top]

Proximity Searching

Proximity searching helps you to specify how near/distant you want your search terms to be. You can specify the distance or nearness of the terms searched as whole numbers. The proximity operator in FreePatentsOnline is "~". The proximity operator should be followed by a whole number indicating the proximity of the search terms. For example, "~10" means within ten words of each other. Note: Ensure that your keywords are enclosed in inverted commas, and there is no space left between the inverted commas and the proximity operator.

Query Results list would have
"Enzyme peptide"~3 Documents with both enzyme and peptide, where both are within 3 words of each other.
"fermentation bacteria milk"~5 Documents where the words fermentation, bacteria and milk, are present within 5 words of each other.

Start here if you already know to search at the USPTO

The search language used by is extremely similar to that of the USPTO. This is no accident. The search at the USPTO is fairly robust (although subject to some limitations, which we will discuss momentarily). It allows keyword searching across a large number of fields, date range searching, and it allows the combining of multiple search criteria with complex Boolean logic. In addition, the use of a similar syntax to the USPTO's search makes the learning curve to use our search engine trivial for the many IP professionals already familiar with the USPTO syntax. For users already familiar with the USPTO's search syntax, we briefly describe the added features of our search.

Differences from the USPTO:

In general, we have designed our search language so that a search that works at the USPTO will work on our search engine (this may not be true for some of the more rarely-used fields at the USPTO -- a complete list of the fields we support can be found below, and on the search page). However, while a search constructed using the USPTO's language will generally work on our search engine, if you limit yourself to using just the USPTO syntax you will be missing out on a great deal of the power of our search engine. Luckily, learning to use all the extra power of the FreePatentsOnline search engine only requires knowing about two extra commands, plus a button on the advanced form.


The Button: Word Stemming

Usage example: With word stemming on, 'metal' will also locate other forms of the word, including 'metals', 'metalized', and 'metallic'. Word stemming is a method of determining the root of a word, and then all possible variants. This is accomplished with a sophisticated formula called the Porter Stemming Algorithm. When it comes to the technical terms often required to express patent concepts, no dictionary even comes close to containing all the words that would be needed. So, rather than rely on an incomplete dictionary, the Porter Stemming Algorithm uses the rules of English (our apologies to those searching non-English documents, this function only works for English at this time) to find words that contain common roots. Now you don't have to miss documents because you were looking for a "metal widget" while the inventor chose to call it a "metallic widget." (Yes, you could use Boolean logic to construct the query "metal OR metals OR metallic OR metalized" but do you really want to do that every time? And are you sure you thought of all the forms of the word? Really sure?). Note that the effect of word stemming on the number of results is fairly significant. The statistics will vary from word to word, but for instance, a search on the word "metal" with word stemming on returns about 16% more results than with word stemming off. You certainly do not want to miss 16% of the potentially relevant documents in your prior art searches! Word stemming is set to "on" by default, and can be turned off by clicking the appropriate button under the search box.


The '~' Command: Proximity Searching

Usage example: "aluminum widgets"~10. Proximity searching is a way to find words that are near each other, rather than a literal phrase in which the words must be right next to each other. Unfortunately for patent searchers, it's easy to say the same thing many different ways -- even when using the same key words. What if you want to find patents about aluminum widgets? You would rather not search for 'aluminum AND widget', because with both "aluminum" and "widget" being very common words (ok, widget probably isn't really that common, but you get the point), that search could return far too many patents to look through in a reasonable amount of time. You could search for the phrase "aluminum widget?" (with the '?' allowing for widget OR widgets), but that is going to miss a ton of patents. What if the inventor said "widgets made of aluminum", or "aluminum may be used to make said widgets", or "the widgets may be made from any of the group comprised of aluminum, bronze, steel, titanium or copper"? You wouldn't find any of those by searching for a literal phrase. That's where proximity searching comes to the rescue. The query "aluminum widget"~10 would find every one of those, while at the same time excluding documents where aluminum and widget were so far apart that they probably had nothing to do with each other. '~' finds words that are in proximity to each other, and you can control how close the two words must be to count as a successful match. The number that follows the '~' indicates how many words are allowed to be between the two keywords. We suggest that 10 is a good distance to start with, but you may find that other values work better for your needs.


The '^2' Command: Search Term Weighting

Usage example: cat^2 OR dog. You may know that some of the search terms you are using are more important than others. If that is the case, the '^' command lets you increase the relative weight of a particular word. Then, the relevancy ranking formula uses this information to return the most accurate results at the top of the list. The caret "^" should be followed by a whole number that indicates the relative importance increase. For example, "^2" means that the word is twice as important as a word with no caret, while "^3" means three times as important, etc.


The '*' and '?' Command: Wildcard Searching

Usage example: electron*, su?. Wildcards are used to match patterns. The single character wildcard '?' represents a single character query. (e.g. su? will search for words like sub, sun, sum etc). The string wildcard '*' represent multiple character queries. (e.g. medic* will search for words like medic, medicine, medicant, medicinal, medicate, etc). The USPTO uses '$' for wildcard searching instead of '*'. However '$' should be the last character in the word. allows you to position the wildcards anywhere inside a word, except at the beginning. You can also combine '*' and '?' (e.g. dep?n* would search for words like depend, dependent, dependence, depending etc). Wildcard searches that create too many word matches may sometimes result in a "search-too-broad" error. It is therefore advisable to be more specific when using wildcards. (e.g. elec* would search for words like elect, elects, electron, electrons, electric, electrical, electricity etc. If you are only interested in patents dealing with electronic, use "electron*" to eliminate the other words. Note: Word stemming should be OFF when using wildcard searches.


Summary of Differences from USPTO Syntax:

Once you understand word stemming, proximity searching, search term weighting and wildcard searching you have mastered the main differences between the USPTO and our site. These search abilities are deceptively simple -- don't underestimate their power to find just what you want to find.

[Back to top  Top]

Start here if you already know how to search at Delphion

The main difference between Delphion and this site is that we follow the USPTO's convention of specifying a field name by putting it in front of the word, followed by a slash, like so: IN/smith. That means find the word "smith" in the Inventor's Name (IN) field. The equivalent command when using Delphion would be "smith IN". That's really the only change you need to make, except of course be sure to use our field name abbreviations. Also, make sure you understand our syntax for proximity searching and search term weighting.