You don't want "teaspoons", your answer is a weight (or volume of a solution).

Hypothetically...

1. Get formula weight of HCO3- = 61.02 grams/mol

2. Figure out how many moles you need;

   M (molarity) = n (moles)/ L (volume)

   0.03303 = n / 1L

   n = 0.03303 moles

3. Convert moles to grams via formula weight;

   0.03303 moles * 61.02 g/mol = 2.02 grams of HCO3- (which is not NaHCO3)

   So then, 2.02 grams of HCO3- in 1 liter of water gives you a 33.03 mM solution (millimolar).

Edit: this just illustrates the math using HCO3-, which you can't do as Matt points out. They must use other salts to get different molarities of sodium (Na) and bicarbonate (HCO3-) ions. At the very least, you should have the whole paper before proceeding, which will include specific Materials & Methods. If you wanted to follow the paper, you really should follow it directly, as in an exact duplication of the study. However, these things tend to balance out over time and this paper is from 1990. If baking soda was the cure for hypertension, I think word would be out (I'm not a physician, and never stay at Holiday Inn Express).

Sodium bicarbonate NaHCO3 has a molar mass of 84 mg per mmol. Sodium bicarbonate is a 1:1 ratio of sodium cations to bicarbonate anions, thus you can't get 6.2 mM sodium with 33 mM HCO3 without some other cation balancing the bicarbonate.

Is the result from that paper really that significant that you want to try it?

Not an answer because I hated chemistry, but thank you for being cautious about baking soda. Many 'a elderly folk have come into the ED after "sody" overdoses in cardiac arrest.