# What is the ANO-RCC basis set and how does it work?

On the basis set exchange, the ANO-RCC basis looks huge, even for hydrogen:

H    S
188.6144500000           0.00096385            -0.0013119              0.00242240            -0.0115701              0.01478099            -0.0212892
28.2765960000           0.00749196            -0.0103451              0.02033817            -0.0837154              0.09403187            -0.1095596
6.4248300000           0.03759541            -0.0504953              0.08963935            -0.4451663              0.53618016            -1.4818260
1.8150410000           0.14339498            -0.2073855              0.44229071            -1.1462710             -0.6089639              3.0272963
0.5910630000           0.34863630            -0.4350885              0.57571439             2.5031871             -1.1148890             -3.7630860
0.2121490000           0.43829736            -0.0247297             -0.9802890             -1.5828490              3.4820812              3.6574131
0.0798910000           0.16510661             0.32252599            -0.6721538              0.03096569            -3.7625390             -2.5012370
0.0279620000           0.02102287             0.70727538             1.1417685              0.30862864             1.6766932              0.89405394
H    P
2.3050000000           0.11279019            -0.2108688              0.75995011            -1.4427420
0.8067500000           0.41850753            -0.5943796              0.16461590             2.3489914
0.2823620000           0.47000773             0.08968888            -1.3710140             -1.9911520
0.0988270000           0.18262603             0.86116340             1.0593155              0.90505601
H    D
1.8190000000           0.27051341            -0.7938035              1.3082770
0.7276000000           0.55101250            -0.0914252             -2.0210590
0.2910400000           0.33108664             0.86200334             1.2498888
H    F
0.9701090000           1.0000000
END


It looks very different from the basis sets that I'm used to, which are Dunning's correlation consistent basis sets. Since the above example is a 4-zeta (QZ), I provide the cc-pVQZ below:

H    S
82.6400000              0.0020060
12.4100000              0.0153430
2.8240000              0.0755790
0.7977000              1.0000000
0.2581000              1.0000000
0.0898900              1.0000000
H    P
2.2920000              1.0000000
0.8380000              1.0000000
0.2920000              1.0000000
H    D
2.0620000              1.0000000
0.6620000              1.0000000
H    F
1.3970000              1.0000000
END


I do not understand why:
1) ANO-RCC has so many more exponents than cc-pVQZ at each angular momentum (2 more for S, 1 more for P, one more for D, but the same number for F!).
2) ANO-RCC has so many more contraction coefficients (3 columns for S, rather than 1 column in cc-pVQZ, and many more non-unity contraction coefficients for S, P, and D, but not F!).

It also appears that the ANO-RCC basis set has to be used in a special way. For example, do I just plug these exponents and contraction coefficients into my program? What if I want a 3-zeta ANO-RCC basis set: I know to chop off the F line, but how many lines for the other exponent types, do I chop off? Can I extrapolate to the CBS limit using 2-zeta, 3-zeta and 4-zeta?

I understand that a good place to look would be the initial papers where ANO-RCC was introduced (by Bjorn Roos?) and the follow-up papers where they reported the optimization of new basis sets. But I do not have access to journal articles since I am not enrolled at a university or working for any company.

• – Tyberius Mar 28 '19 at 0:26