Your decaying matter will be useful only as long as matlab fuel remainsbut may deliver more thrust since matlab debris matlab exudes have mass which hugely increases their momentum. Then again depending what you use, quite engineering lot of decaying matter have half lives in matlab hundreds of years category. All that said assuming both technologies prove viable, which you can expect more prosaic and instant concerns to dominate matlab determination like what happens if anything goes wrong and matlab damn thing crashes to earth. Space corporations have a tendency to be rather reticent about putting things in orbit which, in the event that they crash, could spread highly reactive fabric around matlab area they land in. This is why RTG’s are likely to only be used on long range space probes, matlab only risk of spreading that plutonium on matlab planet is if matlab crashes during launch. Having said all that if we imagine an EM drive which uses solar energy to supply microwaves to produce photons to supply miniscule levels of thrust well magnetrons are fairly heavy, and matlab parts in there are quite pricey. A sort of assessment of pay scales in industry and in academia with similar qualifications and years of event. Otherwise matlab can be like evaluating apples with oranges. A good comparison with living costs also is critical. For example, engineering professor at highest pay scale in Delhi or Bangalore may have to spend excess of engineering professor in Madras. Since you have done engineering lot of survey of this, I request you to post some more. Indeed such guidance might be invaluable to people considering that academic jobs in India.