A key issue regarding the speciation of Al3+ in serum is how well the ligands citric acid and phosphate can compete with the iron transport protein serum transferrin for the aluminum. Previous studies have attempted to measure binding constants for each ligand separately, but experimental problems make it very difficult to obtain stability constants with the accuracy required to make a meaningful comparison between these ligands. In this study, effective binding constants for Al-citrate and Al-phosphate at pH 7.4 have been determined using difference UV spectroscopy to monitor the direct competition between these ligands and transferrin. The analysis of this competition equilibrium also includes the binding of citrate and phosphate as anions to apotransferrin. The effective binding constants are 10(11.59) for the 1:1 Al-citrate complexes and 10(14.90) for the 1:2 Al-citrate complexes. The effective binding constant for the 1:2 Al-phosphate complex is 10(12.02). No 1:1 Al-phosphate complex was detected. Speciation calculations based on these effective binding constants indicate that, at serum concentrations of citrate and phosphate, citrate will be the primary low-molecular-mass ligand for aluminum. Formal stability constants for the Al-citrate system have also been determined by potentiometric methods. This equilibrium system is quite complex, and information from both electrospray mass spectrometry and difference UV experiments has been used to select the best model for fitting the potentiometric data. The mass spectra contain peaks that have been assigned to complexes having aluminum:citrate stoichiometries of 1:1, 1:2, 2:2, 2:3, and 3:3. The difference UV results were used to determine the stability constant for Al(H(-1)cta)(-), which was then used in the least-squares fitting of the potentiometric data to determine stability constants for Al(Hcta)(+), Al(cta), Al(cta)(2)(3-), Al(H(-1)cta)(cta)(4-), Al-2(H(-1)cta)(2)(2-), and Al-3(H(-1)cta)(3)(OH)(4-).