Only if you have done all the research needed to ensure it has no unwanted side effects. Developing safe and effective biological control is the best long-term solution for single species control and has been used effectively in countries like Australia and New Zealand, with the quarantine facilities to do the testing, but it does not always work even then. Chemical control is safer than an untested biological control agent.
While I agree that biological control is preferable, it isn't always possible to achieve the results required due to a range of factors many authors have explored. Sometimes a suitable agent cannot be found even after extensive searches and testing.
There are a number of woody weed killing herbicides being used on Australian rangelands for plants such as Vachellia nilotica, Prosopis spp., Parkinsonia aculeata, etc. Products include Access herbicide and Starane Advanced Herbicide. These are considered safe if used according to their registered label directions. Google search to see active ingredients.
Residual herbicides such as tebuthiuron, while they may incur off-target damage if used incorrectly, are also very useful (though care needed).
In one of my recent projects, I and my colleagues examined new approaches to management of Vachellia nilotica in Australia. You can access a Story Map overview here https://qgsp.maps.arcgis.com/apps/MapSeries/index.html?appid=2ede10d341c8438c98ebc3305499055d
Bioherbicides Australia have been investigating naturally occurring fungal pathogens as woody weed control tools and have now registered Di-Bak Parkinsonia. See https://www.bioherbicides.com.au/bha-weed-control/parkinsonia/