Tricks of the Trade: Searching the unsearchable

Patent searching is already black magic, wrapped in a puzzle, inside an enigma, passing through a black box, and thinking about how something could be unsearchable is an extra layer you just don’t need, so, how do you find something you can’t search for?


To assist, I’ll define what I consider unsearchable.  It’s essentially numerical ranges and relative terms.

Here’s claim 1 of US20200060950:

Carbonate- and magnesium-substituted hydroxyapatite having a part of calcium substituted with magnesium and a part of a phosphate group substituted with a carbonate group, and being composed of primary particles in a non-aggregated state, the primary particles having a particle size of 5 nm or more and 60 nm or less.

I chose this one for the last line: ‘a particle size of 5 nm or more and 60 nm or less’.  It exemplifies both of the unsearchable categories.  It’s essentially a numerical range of 5 to 60 nm, but it also includes the relative terms ‘or more’ and ‘or less’.

Other relative terms you might see are ‘greater than’, ‘less than’, ‘higher’, ‘lower’, ‘heavier’, ‘narrower’ and so on.  You learnt all of these before you were five years old, and they are back.

I summarised the numerical range as 5 to 60 nm, but I could have just as easily written 5 nm to 60 nm, or 5nm to 60nm, or 5-60 nm, and variations on that theme.  Measuring particle size in nanometres may be convenient for this inventor, but others might choose a different unit, so 50 to 600 Angstroms or 0.005 to 0.06 microns, or even 5 x 10-3 to 6 x 10-2 microns is an option.  A search conducted for 5 nm would miss almost all, if not all, of these alternatives.

Similarly a range such as 3 nm to 75 nm would not be located in a specific numerical search for 5 nm to 60 nm even though it would be a highly relevant result.  Considering relative terms again, values of ‘less than 100 nm’ or ‘greater than 2 nm’ would not be located.

So, going full circle, how do you find something you can’t search for?

Searching is a matter of filtering the entire universe of patent applications to get a searchable set you can work with.  I’ve written about that before.  That article considers filters you can apply through a patent database.  They are fairly concrete and well defined, or at least you as the searcher can define them to your satisfaction.

Numerical ranges and relative terms are much less defined, and although you can search for particular numbers, with or without a relative term, you are taking stabs in the dark if you use a patent database filter, and are unlikely to be successful.

At this point you are going to have to apply an internal filter.  No, not the sweary one; a different one.  As you consider each one of your search results (that you obtained with patent database filters) you can apply a mental filter that is equivalent to what you might have liked to apply in that patent database.

The advantage of this is that your mental filter can cope quite well with different numbers defining ranges, or differing units, or various relative terms, something a computer doesn’t do.

So, using the example claim above, a likely search will focus on non-aggregated hydroxyapatite particles with a size, possibly with reference to it being substituted with carbonate or magnesium, although that last bit could be very broad, but the point is that you have not considered at all what that particle size is.

I’ve searched (very quickly and simply, in the title, abstract or claims in PatBase) for (TAC=(hydroxyapatite) AND TAC=(nonaggregated OR (non w1 aggregated)) AND TAC=(particle w2 size*)).

Just looking at the first few I find phrases in the first claim such as:

  • an inorganic fine particle having an average particle size of 0.01.about.5 mu m
  • hexagonal primary particles and having a secondary particle size from about 50 nm to about 5000 nm
  • non-aggregated nano-sized particles,

all of which can be filtered mentally as fitting the 5 to 60 nm range in some way, including the last one which requires some digging to find a range in claim 4.

You can see that none of them would have been located had the particle size formed part of the search string, whether that included the numbers 5 or 60, or as a range, or by using greater than 5 or less than 60, and so on, possibly with the exception of the last one where claim 4 describes a range of 5 nm to 500 nm, but that all depends on how you framed the particle size search string.

The first phrase has different units; ‘mu m’ or micrometres or microns, not nanometres.  The second phrase does use nanometres but the numbers used in the range are different.  The third phrase doesn’t use numbers or units at all, but the more general ‘nano-sized’.

To reiterate, the advantage of this strategy is that your mental filter can cope quite well with different numbers defining ranges, or differing units, or various relative terms, something only a necktop computer can do.