May be related to *ṣ̂ḥk/ṣ̂ḥḳ ‘to laugh’, though hard to explain phonetically, cf. SED I 325. šeḫēḳu ‘to sneeze’ is also probably unrelated (Kogan 2015:98)
Cf. also the PCS root *ṣyḥ in spite of the semantic and phonological problems (Kogan 2015:104).
It comes so close to Aramaic ṣəbūtā both formally and semantically that an Akkadism (or at least some kind of Akkadian influence) is to be seriously considered (see Kogan 2015:427).