Originally from Tartoff, et al. (1987), printed in Cold Spring Harbor Protocol.
Characterized with LB in Danquah & Forde (2007).  

Plasmid yield is, Tartoff et al. say, ≈10  ≈10 µg/mL , ≈3.7-fold for pUC19, fourfold more than LB, but at 65% the medium cost per amount plasmid, according to Danquah & Forde.
Shyam finds that E. coli TG1 saturates at OD₆₀₀ ≈ 18–20 , whereas in LB, in and at OD₆₀₀ ≈ 6 in LB.

Terrific Broth

• 12 g/L tryptone (1.2% m/V)
• 24 g/L yeast extract (2.4% m/V)
• 4 mL/L glycerol (4.34 mM, 0.4% V/V)
• 17 mM KH₂PO₄, 2.31 g/L
• 72 mM K₂HPO₄, 12.54 g/L
• (opt.) 15 g/L agar (1.5% m/V)
• (opt.) 5–10 mM 1 M MgSO₄ or MgCl₂ (nontraditional)
  

It was originally suggested to separately autoclave phosphate buffer and agar. It was recently found that phosphate and agar react to form H₂O₂, but E. coli are less, if at all, significantly impacted due to catalase expression. Commercial mixes do not separate the two, yet instruct autoclaving. There is also reason for separately autoclaving the glycerol, as glycerol partakes in Maillard reactions with tryptone, though probably less so than sugars do. Maillard reactions sequester nutrients and form products that decompose into growth-inhibitory compounds.

Magnesium is deficient in LB, but high level of trace mineral, including Mg²⁺, contamination of phosphate salts might provide enough in TB. To be sure, 10 mM MgSO₄ or MgCl₂ can be added, but is not part of the traditional TB recipe.

Recovery Medium:

Shyam rationalizes that TB supplemented with 0.4% v/v glucose V/V glucose and MgCl₂ and/or MgSO₄ to a final 10–20 mM would be just as good as SOC for transformation recovery. SOB/SOC, has 10 mM MgCl₂ and 10 mM MgSO₄ to help rebuild the cation-associated E. coli LPS.

Procedure for 1 L

1. Dissolve and autoclave on liquid cycle, 20 min, 15 psi:
   1. 12 g tryptone
   2. 24 g yeast extract
   3. 4 mL glycerol (or 8 mL 50% glycerol)
      (or add after autoclaving) 
   4. (opt.) 15 g /L agaragar
   5. (opt.) 5–10 mL 1 M MgSO₄ or MgCl₂
   6. final 900 mL diH₂O
2. Prepare or obtain 100 mL 10× TB buffer (0.17 M KH₂PO₄ 0.72 M K₂HPO₄)
   Dissolve and autoclave on liquid cycle, 20 min, 15 psi:
   1. 2.31 g KH₂PO₄
   2. 12.54 g K₂HPO₄
   3. initial 90 ≈90 mL diH₂O, adjusted to 100 mL final
3. When 900 mL nutrient solution has cooled ≤60°C, add 100 mL sterile buffer.
4. Add any antibiotics when 50–55°C for agar before pouring, or even cooler for liquid.