I don't think anyone can give you a firm answer. It's a complicated and subjective issue from the start. Then if you write about slavery at any length you seen find that you need at least a couple of alternative terms so you don't sound completely monotonous and redundant.

For the most part, people of African descent would be called "black" or "negro" at the time, in print and in formal situations. There's also the other word, but that appears to have been much more often spoken than written down. Histories written up into the 1970s still used Negro -the routine capitalization is a newer usage from the mid-twentieth century- and then it rapidly falls off. This has entirely to do with prevailing mores at the time of writing, though you do still see historians sometimes characterize or paraphrase period ideas using period terms. Doing this well is difficult because one is using words that are visceral punches for many readers and recapitulate historical and recent injustices.

You can use African from context when speaking of enslaved people in the era, many of whom were genuinely born on the continent and endured the Middle Passage. When it's necessary to distinguish between Africans in the New World, who are almost all enslaved, and those in Africa who may or may not be, historians tend to say something like "enslaved Africans" or "African\black slaves". (Not literally with the slash, but as distinct options.) When speaking people in a specific locale, they might be black X, with X being the location as in "black Virginians". African-American sees some use too, particularly by historians integrating the black experience and highlighting black contributions to American history. It's not ideal, but we also disambiguate between whites in the metropole and in the colonies roughly the same way.

Simply calling them "slaves" also usually does the trick. There are also enslaved Native Americans, who share space and time with colonial slavery, but they're badly understudied and American slavery studies usually ignore them so they don't need to distinguish. This is something slavery studies ought to be better about, but we're not there yet. :( More recent historians also tend to prefer "enslaved people" or "enslaved" to "slave" because it shifts the emphasis from putting slavery on as a quality of the victim to highlighting that their status was a choice imposed on them by others, who we then call enslavers. I think that in a generation that will probably be the going thing but right now it's still the new trend.

Most of this would not be technically correct or period-accurate, especially the last part, but as I said a few paragraphs ago we write in our own times and for readers in them and after. It's nice if the terms we use can neatly fit into the era, but period terminology can obscure and mislead as well as any academic jargon. Our purpose as historians is to understand, empathize, and educate so even if the words used back then aren't heinous to us it can be best to set them aside.

I don't write about colonial era slavery often, but in the nineteenth century I tend to use "black Americans" and "black people" as general terms. Those who are enslaved are mostly "the enslaved" or "enslaved people", occasionally slaves. I also make an effort where it makes sense to also pair them up with similar descriptions of whites. If I say "black people" and I need to refer to whites soon after, I'll say "white people" for them, etc. That's roughly in line with what I read in recent works.