How would you even double-blind for this? Double blind means neither the participants nor the people running the study knows who is on what.
You give half the people Soylent and the other half -- what?
You give half the people Soylent and the other half -- what?
Answering your question directly, you could give the control-group a Solyent-type drink that is devoid of nutrients, and to the other group you can give a true Solyent drink. This can test if people report feeling better simply because they are drinking something healthy as opposed to the improvement being caused by the nutrients in the drink.