To ensure confidentiality so that only the intended recipient can decrypt an email, the sender must encrypt in a way that only the recipient can reverse. In public key cryptography, that means encrypting with the recipient's public key. The recipient is the only party who should possess the matching private key, so only they can decrypt the ciphertext. This pattern is fundamental to PKI-based secure email systems such as S/MIME and OpenPGP: the sender looks up or is provided the recipient's certificate/public key, encrypts the message (often by encrypting a randomly generated symmetric session key with the recipient's public key), and the recipient uses their private key to recover the session key and decrypt the content. Encrypting with the sender' s private key would not provide confidentiality; it resembles signing because anyone with the sender's public key could "decrypt" it. Encrypting with a private key of the recipient is also incorrect because private keys are not shared and should never leave the recipient's control. Therefore, the correct key to encrypt the message so only the friend can decrypt it is the recipient's public key.

The Caesar cipher is the classic substitution cipher that encrypts by shifting letters of the alphabet by a fixed number of positions (e.g., shift by 3: A#D, B#E, etc.). It is a monoalphabetic cipher because a single shift value is applied uniformly across the entire message, making it simple and vulnerable to frequency analysis and brute force (only 25 meaningful shifts in the Latin alphabet). Vigenere also involves shifting, but it uses a repeating keyword to vary the shift per character (polyalphabetic), whereas the question's phrasing typically points to the fundamental "shift cipher," which is Caesar. SHA-1 is a cryptographic hash function, not a cipher. Bifid is a fractionation cipher combining Polybius square coordinates and transposition, not a direct shifting method. Therefore, the cipher that uses shifting letters of the alphabet for encryption is the Caesar cipher.

3DES (Triple DES) commonly uses an Encrypt-Decrypt-Encrypt (EDE) sequence. In the two-key form, it encrypts with key K1, decrypts with key K2, then encrypts again with K1. In the three-key form, it encrypts with K1, decrypts with K2, then encrypts with K3. The EDE construction was chosen partly for backward compatibility: if K1=K2=K3, the scheme reduces to single DES, allowing older systems to interoperate in constrained ways. AES and IDEA do not use an EDE triple-stage process as their defining structure; they are single-pass block ciphers with internal rounds. DES is a single-pass algorithm (one key) rather than a triple application with multiple keys. Therefore, the algorithm described-encrypt with one key, decrypt with another, encrypt with the first-is 3DES. Although now considered legacy, it remains a classic example of increasing effective security by applying a block cipher multiple times with independent keys.