It would be stupid to split the RX and TX into separate modules since that would require the manufacturer to duplicate functions. If you are the only person in the world requesting a TX-only module, then the manufacturers must concentrate on builting as cheaply as they can exactly the product the majority of the customers requests.

If someone built a TX-only module, it would be more expensive than a RX+TX module because of the tiny market for it. So a complete two-way module would be the cheapest you can get your handls on a Wifi module unless you get one as a gift or steals one or are lucky and finds one laying on the street.

But a Wifi module isn't just RF electronics. It is also software to control it. And that software would not work with a single-direction module. And you would find that if you did try to just use the transmit side of a module, you would not manage any real bandwith at all without spending months or years of programming to get it to adaptively try to adjust to the environment. But wait a minute - that adaptivity would require that you could hear other traffic in the air. Ooops, your TX-only module wouldn't be able to, so you would end up with a complex 19200 baud one-way modem :)

As has been requested at least twice: Do explain exactly what you want to do. I assume that you need high bandwidth in one direction, but what is the problem with getting the same amount of bandwidth in the other direction for the same amount of money? Or maybe yuor application would be suitable for a laser link? or any other transfer technology there is out there?

All we know is that you are looking for a product that will not solve the problem you want to solve, what ever that problem is. Wifo, Bluetooth, ZigBee, ... is not just radio transmissions. It is also protocols and analysis and adaptive behaviour. All encapsulated into electronics that makes a lot of these decisions based on reception of transmissions from the other side and from other transmitters within range.