Array with primes

[Eric Angelini]

Hello SeqFan,
consider this array where the 1st column is made of all
prime numbers, and the nth term of a row is 2 times the
previous one on the same row [P(n)=P*2^n]:

 p   px2  px4 px8 px16 px32 px64 px128 px256 px512
  2   4    8   16  32   64   128  256   512   1024 ...
  3   6   12   24  48   96   192  384   768   1536 ...
  5  10   20   40  80  160   320  640  1280   2560 ...
  7  14   28   56 112  224   448  896  1792   3584 ...
 11  22   44   88 176  352   704 1408  2816   5632 ...
 13  26   52  104 208  416   832 1664  3328   6656 ...
 17  34   68  136 272  544  1088 2176  4352   8704 ...
 19  38   76  152 304  608  1216 2432  4864   9728 ...
 23  46   92  184 368  736  1472 2944  5888  11776 ...
 29  58  116  232 464  928  1856 3712  7424  14848 ...
 31  62  124  248 496  992  1984 3968  7936  15872 ...
 37  74  148  296 592 1184  2368 4736  9472  18944 ...
 41  82  164  328 656 1312  2624 5248 10496  20992 ...
 43  86  172  344 688 1376  2752 5504 11008  22016 ...
 47  94  188  376 752 1504  3008 6016 12032  24064 ...
 53 106  212  424 848 1696  3392 6784 13568  27136 ...
 .. ...  ...  ...

1) The missing integers in this array are :
   1 9 15 18 21 25 27 30 33 35 36 39 42 45 ... 
   ...which (except the first term 1) are A093642.

2) The main diagonal (NW-SE) reads:
   2 6 20 56 176 416 1088 2432 5888 14848 ...
   ...which is not in the OEIS

3) Other diagonals (NE-SW) sum up to:
    2 = 2
   (4+3) = 7
   (8+6+5) = 19
   (16+12+10+7) = 45
   (38+24+20+14+11) = 101
   (64+48+40+28+22+13) = 215

   ... this is not in the OEIS .

4) If one keeps only the "primitive terms" of A093642 
   one gets:

    A093642 : 1 9 15 18 21 25 27 30 33 35 36 39 42 45 49 50
"primitives": 1 9 15 .. 21 25 .. .. 33 35 .. 39 .. .. 49 ..

... and again, this is not in the OEIS :
    1 9 15 21 25 33 35 39 49 ... 

If this is of interest, could someone kindly check, 
compute more terms and enter those sequences in the OEIS?

Best,
Angelini & Wajnberg